OpenAI Launches GPT-4.5: Advancements and Limitations in AI Technology

ChatGPT-4.5 Review for Authors and Writers : Revolutionary AI or Overpriced Letdown?

It's no secret that AI writing tools have become an integral part of how we create content, brainstorm ideas, and even tackle professional writing tasks. Whether you're a writer, marketer, or just someone curious about the latest tech, you've likely heard the buzz around OpenAI's ChatGPT-4.5. Promising improved creativity and conversational prowess, this latest iteration has sparked plenty of excitement -- but also some skepticism. After all, with so many competitors in the game, from Claude 3.7 Sonnet to Grok 3, the real question is: does ChatGPT-4.5 live up to the hype, or is it simply another overhyped release? If you've ever wrestled with an AI tool that just couldn't quite deliver what you needed -- be it a cohesive story outline, a polished article, or even a simple logline -- you'll know how frustrating it can be. And while OpenAI's latest model does bring some strengths to the table, such as a conversational tone for email writing, its performance in creative writing tasks leaves a lot to be desired. In this overview, the Nerdy Novelist carries out an evaluation of ChatGPT-4.5's wrtiting capabilities, comparing it to its competitors and breaking down its strengths, weaknesses, and overall value. If you're wondering whether this tool is worth your time (and money), you're in the right place. ChatGPT-4.5 demonstrates inconsistent performance in creative writing tasks, often failing to meet the expectations of users seeking high-quality outputs. When tested on tasks such as generating loglines, outlines, and prose, the model frequently produced results that were vague, repetitive, or lacked originality. For example: In comparison, competitors like Claude 3.7 Sonet consistently delivered more engaging and professional-grade content. Even older models, such as GPT-4.0 Creative Writing and GPT-3.5 Sonet, outperformed ChatGPT-4.5 in specific creative tasks. These results raise questions about the extent of improvements introduced in this version. When benchmarked against other models, ChatGPT-4.5 falls short in several critical areas. Competitors such as Claude 3.7 Sonet, Grok 3, and DeepSeek have demonstrated superior performance in generating detailed, coherent, and engaging content. Claude 3.7 Sonet, in particular, stands out for its ability to produce professional-grade writing, making it a preferred choice for businesses and professional writers. While ChatGPT-4.5 shows some improvement in conversational tone, it lacks the creativity and depth required for complex writing tasks. For users prioritizing high-quality content, these deficiencies make it less competitive in the current AI landscape. The model's inability to consistently deliver nuanced and engaging outputs further diminishes its appeal. Here are more detailed guides and articles that you may find helpful on using ChatGPT for writing. A detailed evaluation of ChatGPT-4.5's performance across various writing tasks highlights its limitations. The following results were observed: These findings underscore the model's difficulty in producing detailed, high-quality content, particularly for long-form writing tasks. The inability to meet word count requirements or maintain a consistent tone further limits its utility for professional applications. ChatGPT-4.5 does exhibit some strengths, but these are often overshadowed by its weaknesses. These limitations make it challenging for the model to meet the expectations of users who require nuanced, high-quality content. While it may suffice for casual or less demanding applications, its weaknesses hinder its suitability for professional use. One of the most significant drawbacks of ChatGPT-4.5 is its high cost. At $75 per million input tokens and $150 per million output tokens, it is the most expensive model in its category. By comparison, competitors like Claude 3.7 Sonet offer more affordable pricing, making them a more attractive option for businesses and professional writers. The high cost-to-performance ratio further diminishes GPT-4.5's appeal. For users requiring consistent, high-quality outputs, the expense often outweighs the model's limited benefits. This pricing structure makes it less accessible to a broader audience, particularly those with budget constraints. ChatGPT-4.5 offers a context window of 128k tokens, which is comparable to other models but not superior. Its maximum output of 16k tokens is standard but insufficient for large-scale tasks, such as long-form content generation. These limitations reduce its usability for projects requiring extensive detail, coherence, or length. For users seeking to produce comprehensive and detailed content, these constraints can be a significant drawback. The inability to handle larger-scale tasks effectively limits the model's versatility and appeal in professional settings. Despite OpenAI's claims of enhanced creativity and conversational abilities, ChatGPT-4.5 struggles to meet expectations. Its inconsistent performance in creative writing, combined with high costs and output limitations, makes it less appealing compared to competitors like Claude 3.7 Sonet. For professional writers, businesses, and API users, the model's lack of specificity, coherence, and usability outweighs its strengths. While ChatGPT-4.5 may be suitable for casual users or those with limited requirements, it fails to deliver the quality and value needed for more demanding applications. As the field of AI-driven content generation continues to evolve, ChatGPT-4.5 remains a less competitive option, particularly for users seeking high-quality, cost-effective solutions.

ChatGPT-4.5 is Redefining Creativity and Emotional Intelligence

OpenAI's ChatGPT-4.5 represents a significant step forward in the development of large language models, focusing on enhancing emotional intelligence (EQ) and computational efficiency. Unlike its predecessors, ChatGPT-4.5 isn't here to outshine others in coding or logic-heavy tasks; instead, it's built to excel in areas where emotional intelligence and creativity take center stage. But as exciting as that sounds, it comes with its own set of challenges, from steep costs to limited accessibility. So, what makes ChatGPT-4.5 stand out in a crowded AI landscape? At its core, this model is a bold attempt to redefine how we think about AI's role in creative and collaborative spaces. Whether it's crafting compelling narratives, assisting with strategic planning, or simply offering a more empathetic touch in customer interactions, ChatGPT-4.5 promises a new kind of partnership between humans and machines. But is this leap forward worth the price tag and trade-offs? ChatGPT-4.5 builds upon the foundation of GPT-4, introducing notable improvements in computational efficiency and emotional intelligence. These enhancements make it a specialized tool for tasks that prioritize human interaction and creativity over technical precision. Key areas where ChatGPT-4.5 excels include: This focus on creativity and emotional intelligence positions ChatGPT-4.5 as a versatile tool for industries and applications that value human-like interaction and imaginative problem-solving. ChatGPT-4.5 demonstrates a mixed performance profile, excelling in some areas while underperforming in others. Its strengths and weaknesses reflect its specialized design: This performance profile underscores ChatGPT-4.5's role as a specialized solution rather than a universal tool, catering to specific use cases that prioritize human-centric capabilities. Here are additional guides from our expansive article library that you may find useful on ChatGPT-4.5. One of the most significant challenges facing ChatGPT-4.5 is its cost. At $75 per million tokens, it is substantially more expensive than GPT-4, which costs $2.5 per million tokens. This steep pricing reflects the model's resource-intensive architecture, which relies heavily on high-performance GPUs. Additionally, global GPU shortages have delayed its availability to Pro and Plus users, further limiting its accessibility. These factors position ChatGPT-4.5 as a premium offering, making it less viable for cost-sensitive applications or users seeking affordable AI solutions. While its capabilities are impressive, the high costs and resource demands may deter widespread adoption, particularly in competitive markets where lower-cost alternatives are available. ChatGPT-4.5 introduces advancements in both its pre-training and post-training processes, which contribute to its unique capabilities: These innovations highlight ChatGPT-4.5's dual emphasis on creativity and emotional intelligence, setting it apart from models that prioritize technical problem-solving or logical reasoning. Its design reflects a growing trend in AI development toward bridging human-centric and technical capabilities. ChatGPT-4.5 is best suited for applications that prioritize creativity, collaboration, and emotional intelligence. Its primary use cases include: While ChatGPT-4.5 excels in these areas, it is less effective for tasks requiring advanced reasoning, coding, or technical precision. This specialization makes it a valuable tool for specific industries and applications but limits its utility as a general-purpose AI solution. In the rapidly evolving AI market, ChatGPT-4.5 occupies a premium niche. OpenAI's pricing strategy places it well above free models offered by competitors like Google and mid-tier options from companies such as Anthropic. While ChatGPT-4.5 offers unique capabilities, its high costs and limited accessibility may deter users seeking more affordable or versatile alternatives. Additionally, the rise of open-weight models presents a significant challenge to ChatGPT-4.5's market position. These models often deliver competitive performance at a fraction of the cost, making them attractive options for businesses and developers. As a result, ChatGPT-4.5's premium pricing and specialized focus may limit its appeal in an increasingly competitive landscape. Despite its advancements, ChatGPT-4.5 has several limitations that users should consider: These trade-offs reflect ChatGPT-4.5's design priorities, which favor creativity and emotional intelligence over technical versatility. While these limitations may not impact its performance in creative or human-centric applications, they do restrict its utility for users seeking a more comprehensive AI solution. ChatGPT-4.5 serves as a bridge between human-centric and technical AI capabilities, laying the groundwork for future innovations in the field. Its emphasis on emotional intelligence and computational efficiency reflects a broader shift in AI development toward models that prioritize human-like interaction and creativity. However, the model's high costs and resource demands highlight a growing divide in the AI market between premium, mid-tier, and free offerings. As the industry continues to evolve, ChatGPT-4.5's role may shift, influencing the design and deployment of next-generation models. Its introduction underscores both the potential and the limitations of large language models in addressing diverse user needs, offering valuable insights into the future of AI development.

OpenAI GPT-4.5 First Impressions : Game-Changer or Overpriced Upgrade?

OpenAI's GPT-4.5 represents a significant step in the evolution of artificial intelligence, offering enhancements in emotional intelligence, alignment, and multimodal capabilities. These improvements aim to refine its performance across various tasks, making it a more versatile tool for users. However, the AI model has also faced criticism for its incremental advancements and specific limitations, particularly in areas like coding and software engineering. This first impression overview by World of AI provides more insights into GPT-4.5's key features, challenges, and practical applications, helping you determine its relevance to your needs. Explore what makes GPT-4.5 stand out, where it struggles, and how it might fit into your personal or professional toolkit. Whether you're a creative looking for a tool to enhance your projects, a business leader exploring AI for strategic planning, or simply someone intrigued by innovative tech, we'll break down the advancements and limitations of this model in a way that's easy to understand. By the end, you'll have a clearer picture of whether GPT-4.5 is the right fit for your needs -- or if it's better to stick with what you already know. GPT-4.5 builds upon its predecessor with several notable upgrades designed to improve its usability, accuracy, and versatility. If you've worked with earlier versions, here are the standout features: These advancements position GPT-4.5 as a versatile tool for both creative and analytical tasks, offering practical benefits in areas like content creation, strategic decision-making, and visual data interpretation. Despite its improvements, GPT-4.5 is not without its limitations. These shortcomings may affect its suitability for certain users and applications: These limitations suggest that GPT-4.5 is best suited for specific, targeted use cases rather than as a comprehensive, all-purpose AI solution. Find more information on ChatGPT capabilities by browsing our extensive range of articles, guides and tutorials. One of the most significant challenges associated with GPT-4.5 is its pricing model. The cost of processing input and output tokens is notably high, which can deter casual users or smaller organizations. While businesses with advanced application needs may find the investment worthwhile, individual users or startups may struggle to justify the expense. This pricing strategy appears to prioritize enterprise-level clients, potentially limiting broader accessibility and adoption. For those considering GPT-4.5, it's essential to weigh the potential benefits against the financial investment, particularly if your use case does not fully use the model's advanced capabilities. GPT-4.5's development reflects both its strengths and the challenges inherent in creating advanced AI systems. While the model benefits from expanded pre-training and a larger architecture, these advancements come with trade-offs: These constraints highlight the ongoing challenge of balancing performance, scalability, and cost in the development of advanced AI technologies. Despite its limitations, GPT-4.5 excels in several niche applications, making it a valuable tool for specific tasks. Its strengths are particularly evident in the following areas: However, its limited effectiveness in programming and debugging tasks restricts its appeal for developers and engineers, who may find more value in specialized tools tailored to their needs. GPT-4.5 represents a meaningful advancement in AI technology, offering improvements in emotional intelligence, alignment, and multimodal capabilities. These features make it a strong contender for creative and analytical tasks, particularly in niche applications where its strengths can be fully used. However, its high costs, incremental improvements over GPT-4, and underperformance in coding tasks may limit its appeal for certain users. When considering GPT-4.5, it's essential to evaluate whether its capabilities align with your specific requirements. For businesses and organizations requiring advanced AI tools, GPT-4.5 offers significant potential. On the other hand, casual users, developers, or smaller organizations may find its limitations and pricing outweigh its benefits. Ultimately, GPT-4.5 is a promising but imperfect tool, best suited for targeted applications rather than broad, general-purpose use.

GPT-4.5 vs GPT-4.0 : What You Need to Know Before Upgrading

OpenAI's GPT-4.5 represents a significant step forward in artificial intelligence, offering refined conversational abilities, improved emotional intelligence, and a reduced likelihood of generating inaccurate information, commonly referred to as AI hallucinations. These advancements position GPT-4.5 as a versatile tool in the AI landscape. However, questions about its cost-effectiveness and suitability for specialized tasks remain. In this testing carried out by All About AI, reveals what makes GPT-4.5 stand out -- and where it falls short. Providing more insights into GPT-4.5's capabilities, limitations, and potential applications, providing a detailed perspective on its role in the evolving world of AI. Whether you're someone who relies on AI for creative tasks, problem-solving, or just everyday convenience, this breakdown will help you decide if this model is worth the investment. While it promises reduced errors and a more human-like touch, questions about its cost and practical value remain. GPT-4.5 builds upon the foundation of its predecessor, GPT-4.0, with several key enhancements that improve its overall performance and usability. These improvements include: These advancements make GPT-4.5 a robust general-purpose AI tool, capable of handling a wide range of tasks with greater precision and adaptability than its predecessor. While GPT-4.5 offers notable improvements, it is not without its limitations. These challenges may influence its adoption, particularly for users with specific requirements or budget constraints. Key concerns include: These limitations suggest that while GPT-4.5 is a powerful tool for general applications, it may not be the ideal choice for every user or scenario, particularly those requiring specialized capabilities or operating under tight budget constraints. Below are more guides on OpenAI from our extensive range of articles. GPT-4.5 excels in several practical applications, particularly those that benefit from its enhanced conversational abilities and creative problem-solving skills. Its most effective use cases include: However, GPT-4.5 is less suited for workflows requiring advanced technical reasoning, complex coding, or high levels of autonomy. In these cases, its performance gains over GPT-4.0 may not justify the additional cost, and specialized tools may be more appropriate. Safety and ethical considerations remain central to the deployment of GPT-4.5. While the model demonstrates medium risks in areas such as persuasion and potential misuse, OpenAI is actively working to address these concerns. Through iterative testing, updates, and the implementation of safeguards, the company aims to minimize potential harm and ensure responsible use. For users, this means that while GPT-4.5 offers powerful capabilities, it should be employed thoughtfully and with appropriate safeguards in place. Ethical considerations, such as preventing misuse and making sure transparency, are critical as AI technology continues to evolve and integrate into various aspects of society. When compared to GPT-4.0, GPT-4.5 delivers incremental improvements in conversational AI, emotional intelligence, and factual accuracy. However, its performance in specialized tasks, such as domain-specific reasoning or technical problem-solving, remains largely comparable to its predecessor. In these areas, specialized models like Deep Research or other domain-specific AI tools often outperform GPT-4.5. This positions GPT-4.5 as a strong choice for broad, general-purpose applications but less compelling for users with niche or highly technical requirements. Its role as a generalist AI tool makes it versatile, but it may not meet the needs of users seeking innovative performance in specialized fields. The reception to GPT-4.5 has been mixed, reflecting both its strengths and its limitations. On one hand, its advancements in conversational AI and emotional intelligence have been praised, particularly for creative and business-oriented tasks. On the other hand, its high cost and limited differentiation from GPT-4.0 have led some users to question its overall value. Looking ahead, the long-term success of GPT-4.5 will depend on its accessibility, further refinements, and how effectively it addresses user concerns. OpenAI's commitment to iterative improvement and ethical deployment will play a crucial role in shaping the model's future impact. For now, GPT-4.5 represents a meaningful evolution in AI, offering enhanced capabilities for general-purpose tasks while highlighting the ongoing need to balance innovation with practicality.

GPT-4.5 Officially Launches : OpenAI's Latest Leap in Artificial Intelligence

OpenAI has announced the launch of GPT-4.5, its most advanced AI model to date, marking a significant step forward in artificial intelligence. Initially made available to ChatGPT Pro users and developers via API, GPT-4.5 is designed to provide a more refined and efficient user experience. It achieves this through enhanced contextual understanding, improved emotional nuance, and advanced problem-solving capabilities. Building on the strengths of its predecessors, GPT-4.5 addresses critical challenges such as reducing hallucinations and aligning more effectively with human feedback, making sure a more reliable and user-friendly interaction. What makes GPT-4.5 stand out isn't just its technical advancements -- it's how seamlessly it integrates into real-world scenarios. From reducing frustrating inaccuracies (goodbye, hallucinations!) to better understanding context and tone, this model is all about making interactions feel natural and productive. Whether you're a developer, a writer, or just someone curious about the possibilities of AI, GPT-4.5 promises to be more than just a tool -- it's a partner in creativity, problem-solving, and communication. GPT-4.5 introduces a range of advancements that distinguish it from earlier models. By using innovative unsupervised learning techniques, OpenAI has expanded the model's knowledge base and improved its ability to interpret and respond to complex queries. This results in a more accurate, intuitive, and dependable AI system. Key improvements include: These features make GPT-4.5 a versatile tool for a wide range of applications, including technical programming, creative writing, and complex problem-solving. Its ability to adapt to diverse user needs ensures it remains a valuable resource across industries. The development of GPT-4.5 incorporates state-of-the-art training methodologies, including reinforcement learning from human feedback (RLHF). This approach ensures the model aligns closely with user expectations while maintaining scalability and efficiency. OpenAI has optimized its training infrastructure to process vast datasets without compromising performance, resulting in a model that is both powerful and resource-efficient. A key innovation lies in scalable alignment techniques. These methods allow GPT-4.5 to learn effectively from smaller, curated datasets, improving its ability to generate structured outputs and perform advanced reasoning tasks. This focus on alignment ensures the model delivers outputs that are not only accurate but also contextually appropriate, meeting the diverse needs of users. Additionally, GPT-4.5 incorporates advanced reasoning capabilities, allowing it to handle complex scenarios with greater precision. This makes it particularly effective in fields requiring logical analysis, such as legal research, scientific exploration, and technical troubleshooting. By prioritizing alignment and reasoning, OpenAI has created a model that is both robust and adaptable. Take a look at other insightful guides from our broad collection that might capture your interest in OpenAI. GPT-4.5 sets new standards in AI performance, surpassing its predecessors in benchmarks for accuracy, factuality, and creative intelligence. Its ability to handle multimodal tasks -- such as interpreting images and generating text-based insights -- positions it as a leader in the AI landscape. Furthermore, its multilingual capabilities make it a valuable tool for users across the globe, breaking down language barriers and fostering international collaboration. Notable capabilities include: These advancements make GPT-4.5 a reliable and innovative solution for both individual users and enterprise applications. Its ability to adapt to specific use cases ensures it remains relevant in a rapidly evolving technological landscape. GPT-4.5 is designed with user experience at its core, offering features that foster intuitive and collaborative interactions. Tools like file and image uploads, search integration, and the canvas tool enhance its usability, making it a practical choice for professionals, developers, and creatives alike. Whether you're drafting a report, debugging code, or brainstorming ideas, GPT-4.5 adapts to your needs with precision and ease. For developers, the model introduces advanced functionalities such as structured output generation and function calling. These features simplify integration into existing systems, allowing more efficient development processes. Additionally, GPT-4.5's ability to process multimodal inputs ensures it can handle diverse tasks, from analyzing visual data to generating detailed textual explanations. The model's focus on accessibility and ease of use ensures that users of all skill levels can benefit from its capabilities. By combining advanced features with a user-friendly interface, GPT-4.5 provides a seamless experience that caters to both casual users and professionals. GPT-4.5 lays the groundwork for the next generation of reasoning-based AI systems. OpenAI emphasizes the importance of combining unsupervised learning with advanced reasoning capabilities to create more sophisticated models. By encouraging community exploration and feedback, OpenAI aims to uncover new applications and further refine the model. Currently, GPT-4.5 is accessible to Pro users on web, mobile, and desktop platforms. Developers on paid tiers can also use its advanced features. OpenAI plans to expand availability to Plus, team, EDU, and enterprise users in the coming weeks, making sure broader access to its capabilities. This phased rollout reflects OpenAI's commitment to making innovative AI technology widely available while maintaining a focus on user feedback and continuous improvement. As OpenAI continues to push the boundaries of AI development, GPT-4.5 represents a pivotal milestone in the evolution of artificial intelligence. Its combination of advanced training techniques, practical features, and user-focused design sets the stage for the next wave of reasoning-based systems, offering exciting opportunities for users and developers alike.

ChatGPT releases new GPT-4.5: What's new in the more advanced 'emotional' version?

OpenAI has unveiled GPT-4.5, the latest and most advanced iteration of its AI model, designed to refine human-AI interactions with improved emotional intelligence, accuracy, and contextual understanding. Unlike its predecessors, GPT-4.5 prioritises intuitive dialogue, making conversations more natural and engaging. With reduced hallucination rates, enhanced knowledge processing, and faster response times, the model sets a new benchmark for AI communication.The artificial intelligence race has entered a new phase with OpenAI's latest release -- GPT-4.5. Officially launched as a research preview for ChatGPT Pro users, the model will be accessible to Plus and Team subscribers next week. This version represents a major step forward in AI conversations, offering users a chatbot that feels more intuitive, reliable, and human-like than ever before. GPT-4.5 arrives at a time when AI competition is fierce, with models like DeepSeek-R1, Grok 3, and Claude 3.7 Sonnet pushing the boundaries of what AI can achieve. OpenAI's new offering aims to reclaim dominance in large language models (LLMs), reinforcing its commitment to innovation and user experience. While GPT-4o introduced multimodal capabilities and faster processing, GPT-4.5 focuses on refining AI's ability to understand nuance, engage in meaningful conversations, and provide insightful responses. According to OpenAI's Vice President of Research, Mia Glaese, "What sets the model apart is its ability to engage in warm, intuitive, naturally flowing conversations, and we think it has a stronger understanding of what users mean when they ask for something." GPT-4.5's development centres on scaling two complementary AI paradigms: unsupervised learning and reasoning. Unlike reasoning-heavy models such as OpenAI's o1 and o3-mini, which break down complex problems step-by-step, GPT-4.5 generates responses based on learned patterns. This approach enables: However, this also means GPT-4.5 struggles with explicit logic-based reasoning, making it less suitable for complex mathematical or coding challenges that require multi-step problem-solving. OpenAI's internal testing highlights several key areas where GPT-4.5 shines: One compelling example from OpenAI's testing involved the model accurately identifying Claude Lorrain's The Trojan Women Setting Fire to Their Fleet and explaining its literary significance within The Aeneid -- demonstrating both knowledge depth and contextual understanding. Starting today, ChatGPT Pro users can select GPT-4.5 on the web, mobile, and desktop versions of ChatGPT. Plus and Team users will receive access next week. Developers can also experiment with GPT-4.5 via OpenAI's API, which supports function calling, structured outputs, and vision capabilities through image inputs. However, GPT-4.5 does not currently support multimodal features such as voice mode, video generation, or screen sharing -- though OpenAI has hinted at future updates. GPT-4.5 marks the end of OpenAI's non-chain-of-thought models. Future iterations will focus on explicit reasoning, where AI takes time to process and logically structure responses before answering. This shift could redefine AI's ability to solve complex problems, particularly in STEM fields. As AI research moves toward reasoning-based intelligence, OpenAI faces stiff competition from companies like Google and DeepSeek. The industry is now looking beyond simple pattern recognition and towards AI that can logically break down complex tasks -- a direction OpenAI is already exploring with its o1 and o3 models. In the meantime, GPT-4.5 offers a more natural, engaging chatbot experience, setting a new standard for AI-driven conversations. While it may not be the most logic-driven model, it represents a crucial step towards AI that doesn't just respond but truly understands.

OpenAI releases 'largest, most knowledgable' model GPT-4.5 with reduced hallucinations and high API price

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More It's here: OpenAI has announced the release of GPT-4.5, a research preview of its latest and most powerful large language model (LLM) for chat applications. Unfortunately, it's far-and-away OpenAI's most expensive model (more on that below). It's also not a "reasoning model," the new class of models offered by OpenAI, DeepSeek, Anthropic and many others that produce "chains-of-thought," or stream-of-consciousness like text blocks in which they reflect on their own assumptions and conclusions to try and catch errors before serving up responses/outputs to users. It's still more of a classical LLM. Nonetheless, acording to OpenAI co-founder and CEO Sam Altman's post on the social network X, GPT-4.5 is: "the first model that feels like talking to a thoughtful person to me. i have had several moments where i've sat back in my chair and been astonished at getting actually good advice from an AI." However, he cautioned that the company is bumping up against the upper end of its supply of graphics processing units (GPUs) and has had to limit access as a result: "bad news: it is a giant, expensive model. we really wanted to launch it to plus and pro at the same time, but we've been growing a lot and are out of GPUs. we will add tens of thousands of GPUs next week and roll it out to the plus tier then. (hundreds of thousands coming soon, and i'm pretty sure y'all will use every one we can rack up.) this isn't how we want to operate, but it's hard to perfectly predict growth surges that lead to GPU shortages." Starting today, GPT-4.5 is available to subscribers of OpenAI's most expensive subscription tier, ChatGPT Pro ($200 USD/month), and developers across all paid API tiers, with plans to expand access to the far less costly Plus and Team tiers ($20/$30 monthly) next week. GPT‑4.5 is able to access search and OpenAI's ChatGPT Canvas mode, and users can upload files and images to it, but it doesn't have other multimodal features like Voice Mode, video, and screensharing -- yet. OpenAI is hosting a livestream event today at 12 pm PT/ 3 pm ET, where OpenAI researchers will discuss the model's development and capabilities. Advancing AI with unsupervised learning GPT-4.5 represents a step forward in AI training, particularly in unsupervised learning, which enhances the model's ability to recognize patterns, draw connections, and generate creative insights. During the livestream, OpenAI researchers noted how it was trained on data generated by smaller AI models and that this improved its "world model." They also stated it was pre-trained across multiple data centers concurrently, suggesting a decentralized approach similar to that of rival lab Nous Research. This training regimen apparently helped GPT-4.5 learn to produce more natural and intuitive interactions, follow user intent more accurately, and demonstrate greater emotional intelligence. The model builds on OpenAI's previous work in AI scaling, reinforcing the idea that increasing data and compute power leads to better AI performance. Compared to its predecessors, GPT-4.5 is expected to produce fewer hallucinations, making it more reliable across a broad range of topics. What makes GPT-4.5 stand out? According to OpenAI, GPT-4.5 is designed to create warm, intuitive, and naturally flowing conversations. It has a stronger grasp of nuance and context, enabling more human-like interactions and a greater ability to collaborate effectively with users. The model's expanded knowledge base and improved ability to interpret subtle cues allow it to excel in various applications, including: * Writing assistance: Refining content, improving clarity, and generating creative ideas. * Programming support: Debugging, suggesting code improvements, and automating workflows. * Problem-solving: Providing detailed explanations and assisting in practical decision-making. GPT-4.5 also incorporates new alignment techniques that enhance its ability to understand human preferences and intent, further improving user experience. How to access GPT-4.5 Starting today, ChatGPT Pro users can select GPT-4.5 in the model picker on web, mobile, and desktop. Next week, OpenAI will begin rolling it out to Plus and Team users. For developers, GPT-4.5 is being made available through OpenAI's API, including the Chat Completions API, Assistants API, and Batch API. It supports key features like function calling, structured outputs, streaming, system messages, and image inputs, making it a versatile tool for various AI-driven applications. However, it currently does not support multimodal capabilities such as voice mode, video, or screen sharing. Pricing and implications for enterprise decision-makers Enterprises and team leaders stand to benefit significantly from the capabilities introduced with GPT-4.5. With its enhanced reliability and natural conversational abilities, GPT-4.5 can support a wide range of business functions: * Improved Customer Engagement: Businesses can integrate GPT-4.5 into support systems for faster, more natural interactions. * Enhanced Content Generation: Marketing and communications teams can produce high-quality, on-brand content efficiently. * Streamlined Operations: AI-powered automation can assist in debugging, workflow optimization, and strategic decision-making. * Scalability and Customization: The API allows for tailored implementations, enabling enterprises to build AI-driven solutions suited to their needs. At the same time, the pricing for GPT-4.5 through OpenAI's API for third-party developers looking to build applications on the model appears shockingly high, at $75/$180 per million input/output tokens compared to $2.50/$10 for GPT-4o. And with other rival models released recently -- from Anthropic's Claude 3.7 to Google's Gemini 2 Pro to OpenAI's own reasoning "o" series (o1, o3-mini high, o3) -- the question will become if GPT-4.5s value is worth the relatively high cost, especially through the API. Early reactions from fellow AI researchers and power users vary widely The release of GPT-4.5 has sparked mixed reactions from AI researchers and tech enthusiasts on the social network X, particularly after a version of the model's "System Card" (a technical document outlining its training and evaluations) leaked earlier (included below at the bottom of this article), revealing a variety of benchmark results ahead of the official announcement. Teknium (@Teknium1), the pseudonymous co-founder of rival AI model provider Nous Research, expressed disappointment in the new model, pointing out minimal improvements in MMLU (multilingual understanding) scores and real-world coding benchmarks compared to other leading LLMs. "It's been 2+ years and 1000s of times more capital has been deployed since GPT-4... what happened?" he asked. Others noted that GPT-4.5 underperformed relative to OpenAI's o3-mini model in software engineering benchmarks, raising questions about whether this release represents significant progress. However, some users defended the model's potential beyond raw benchmarks. Software developer Haider (@slow_developer) highlighted GPT-4.5's 10x computational efficiency improvement over GPT-4 and its stronger general-purpose capabilities compared to OpenAI's STEM-focused o-series models. AI news poster Andrew Curran (@AndrewCurran_) took a more qualitative view, predicting that GPT-4.5 would set new standards in writing and creative thought, calling it OpenAI's "Opus." These discussions underscore a broader debate in AI: Should progress be measured purely in benchmarks, or do qualitative improvements in reasoning, creativity, and human-like interactions hold greater value? Still in research preview OpenAI is positioning GPT-4.5 as a research preview to gain deeper insights into its strengths and limitations. The company remains committed to understanding how users interact with the model and identifying unexpected use cases. "We're sharing GPT-4.5 as a research preview to better understand its strengths and limitations," OpenAI stated. "Scaling unsupervised learning continues to drive AI progress, improving accuracy, fluency, and reliability." As OpenAI continues to refine its models, GPT-4.5 serves as a foundation for future AI advancements, particularly in reasoning and tool-using agents. While GPT-4.5 is already demonstrating impressive capabilities, OpenAI is actively evaluating its long-term role within its ecosystem. With its broader knowledge base, improved emotional intelligence, and more natural conversational abilities, GPT-4.5 is set to offer significant improvements for users across various domains. OpenAI is keen to see how developers, businesses, and enterprises integrate the model into their workflows and applications. As AI continues to evolve, GPT-4.5 marks another milestone in OpenAI's pursuit of more capable, reliable, and user-aligned language models, promising new opportunities for innovation in the enterprise landscape.

OpenAI releases GPT-4.5 claiming 10X efficiency over GPT-4, but says it's 'not a frontier model'

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More It's here: OpenAI has announced the release of GPT-4.5, a research preview of its latest and most powerful large language model (LLM) for chat applications -- yet pointedly, the company says specifically in the model's system card (a report outlining its capabilities, which leaked early online and is attached to the bottom of this piece) that "GPT-4.5 is not a frontier model, but it is OpenAI's largest LLM, improving on GPT-4's computational efficiency by more than 10x." It's also not a "reasoning model," the new class of models offered by OpenAI, DeepSeek, Claude and many others that produce "chains-of-thought," or stream-of-consciousness like text blocks in which they reflect on their own assumptions and conclusions to try and catch errors before serving up responses/outputs to users. It's still more of a classical LLM. It's also, perplexingly, far-and-away OpenAI's most expensive model (more on that below). Nonetheless, acording to OpenAI co-founder and CEO Sam Altman's post on the social network X, GPT-4.5 is: "the first model that feels like talking to a thoughtful person to me. i have had several moments where i've sat back in my chair and been astonished at getting actually good advice from an AI." However, he cautioned that the company is bumping up against the upper end of its supply of graphics processing units (GPUs) and has had to limit access as a result: "bad news: it is a giant, expensive model. we really wanted to launch it to plus and pro at the same time, but we've been growing a lot and are out of GPUs. we will add tens of thousands of GPUs next week and roll it out to the plus tier then. (hundreds of thousands coming soon, and i'm pretty sure y'all will use every one we can rack up.) this isn't how we want to operate, but it's hard to perfectly predict growth surges that lead to GPU shortages." Starting today, GPT-4.5 is available to subscribers of OpenAI's most expensive subscription tier, ChatGPT Pro ($200 USD/month), and developers across all paid API tiers, with plans to expand access to the far less costly Plus and Team tiers ($20/$30 monthly) next week. GPT‑4.5 is able to access search and OpenAI's ChatGPT Canvas mode, and users can upload files and images to it, but it doesn't have other multimodal features like Voice Mode, video, and screensharing -- yet. OpenAI is hosting a livestream event today at 12 pm PT/ 3 pm ET, where OpenAI researchers will discuss the model's development and capabilities. Advancing AI with unsupervised learning GPT-4.5 represents a step forward in AI training, particularly in unsupervised learning, which enhances the model's ability to recognize patterns, draw connections, and generate creative insights. During the livestream, OpenAI researchers noted how it was trained on data generated by smaller AI models and that this improved its "world model." They also stated it was pre-trained across multiple data centers concurrently, suggesting a decentralized approach similar to that of rival lab Nous Research. This training regimen apparently helped GPT-4.5 learn to produce more natural and intuitive interactions, follow user intent more accurately, and demonstrate greater emotional intelligence. The model builds on OpenAI's previous work in AI scaling, reinforcing the idea that increasing data and compute power leads to better AI performance. Compared to its predecessors, GPT-4.5 is expected to produce fewer hallucinations, making it more reliable across a broad range of topics. What makes GPT-4.5 stand out? According to OpenAI, GPT-4.5 is designed to create warm, intuitive, and naturally flowing conversations. It has a stronger grasp of nuance and context, enabling more human-like interactions and a greater ability to collaborate effectively with users. The model's expanded knowledge base and improved ability to interpret subtle cues allow it to excel in various applications, including: * Writing assistance: Refining content, improving clarity, and generating creative ideas. * Programming support: Debugging, suggesting code improvements, and automating workflows. * Problem-solving: Providing detailed explanations and assisting in practical decision-making. GPT-4.5 also incorporates new alignment techniques that enhance its ability to understand human preferences and intent, further improving user experience. How to access GPT-4.5 Starting today, ChatGPT Pro users can select GPT-4.5 in the model picker on web, mobile, and desktop. Next week, OpenAI will begin rolling it out to Plus and Team users. For developers, GPT-4.5 is being made available through OpenAI's API, including the Chat Completions API, Assistants API, and Batch API. It supports key features like function calling, structured outputs, streaming, system messages, and image inputs, making it a versatile tool for various AI-driven applications. However, it currently does not support multimodal capabilities such as voice mode, video, or screen sharing. Pricing and implications for enterprise decision-makers Enterprises and team leaders stand to benefit significantly from the capabilities introduced with GPT-4.5. With its enhanced reliability and natural conversational abilities, GPT-4.5 can support a wide range of business functions: * Improved Customer Engagement: Businesses can integrate GPT-4.5 into support systems for faster, more natural interactions. * Enhanced Content Generation: Marketing and communications teams can produce high-quality, on-brand content efficiently. * Streamlined Operations: AI-powered automation can assist in debugging, workflow optimization, and strategic decision-making. * Scalability and Customization: The API allows for tailored implementations, enabling enterprises to build AI-driven solutions suited to their needs. At the same time, the pricing for GPT-4.5 through OpenAI's API for third-party developers looking to build applications on the model appears shockingly high, at $75/$180 per million input/output tokens compared to $2.50/$10 for GPT-4o. And with other rival models released recently -- from Anthropic's Claude 3.7 to Google's Gemini 2 Pro to OpenAI's own reasoning "o" series (o1, o3-mini high, o3) -- the question will become if GPT-4.5s value is worth the relatively high cost, especially through the API. Early reactions from fellow AI researchers and power users vary widely The release of GPT-4.5 has sparked mixed reactions from AI researchers and tech enthusiasts on the social network X, particularly after the System Card leak revealed benchmark results ahead of the official announcement. Teknium (@Teknium1), the pseudonymous co-founder of rival AI model provider Nous Research, expressed disappointment in the new model, pointing out minimal improvements in MMLU (multilingual understanding) scores and real-world coding benchmarks compared to other leading LLMs. "It's been 2+ years and 1000s of times more capital has been deployed since GPT-4... what happened?" he asked. Others noted that GPT-4.5 underperformed relative to OpenAI's o3-mini model in software engineering benchmarks, raising questions about whether this release represents significant progress. However, some users defended the model's potential beyond raw benchmarks. Software developer Haider (@slow_developer) highlighted GPT-4.5's 10x computational efficiency improvement over GPT-4 and its stronger general-purpose capabilities compared to OpenAI's STEM-focused o-series models. AI news poster Andrew Curran (@AndrewCurran_) took a more qualitative view, predicting that GPT-4.5 would set new standards in writing and creative thought, calling it OpenAI's "Opus." These discussions underscore a broader debate in AI: Should progress be measured purely in benchmarks, or do qualitative improvements in reasoning, creativity, and human-like interactions hold greater value? A research preview OpenAI is positioning GPT-4.5 as a research preview to gain deeper insights into its strengths and limitations. The company remains committed to understanding how users interact with the model and identifying unexpected use cases. "We're sharing GPT-4.5 as a research preview to better understand its strengths and limitations," OpenAI stated. "Scaling unsupervised learning continues to drive AI progress, improving accuracy, fluency, and reliability." As OpenAI continues to refine its models, GPT-4.5 serves as a foundation for future AI advancements, particularly in reasoning and tool-using agents. While GPT-4.5 is already demonstrating impressive capabilities, OpenAI is actively evaluating its long-term role within its ecosystem. With its broader knowledge base, improved emotional intelligence, and more natural conversational abilities, GPT-4.5 is set to offer significant improvements for users across various domains. OpenAI is keen to see how developers, businesses, and enterprises integrate the model into their workflows and applications. As AI continues to evolve, GPT-4.5 marks another milestone in OpenAI's pursuit of more capable, reliable, and user-aligned language models, promising new opportunities for innovation in the enterprise landscape.

OpenAI Unveils GPT-4.5, Its Largest And Most Sophisticated Chat Model Yet

OpenAI's impact on the AI world cannot be ignored, and it has revolutionized the industry and given the initiatives a new direction. The company behind ChatGPT has been arduously working to make its mark and keep bringing forward innovative products. The efforts have especially escalated with the growing popularity of DeepSeek, and the company gave users a heads-up about the products it has in the pipeline. Following that, there has been a stream of updates, and the tech giant is not slowing down as it has now introduced GPT-4.5, its latest version, and the best model to date. It seems like 2025 is the year for AI models as Chinese-based startup DeepSeek took everyone by surprise with its powerful r1 model, that was remarkably trained at a very low cost. A month later, OpenAI announced its o3-mini, which came with enhanced reasoning capabilities and a better response rate. Now, OpenAI has unveiled its latest generative AI model, which is said to be its largest model yet and very computer-intensive. OpenAI has announced GPT-4.5 which brings more notable improvements in accuracy and has far better emotional intelligence. The company, however, clarified that while the model is powerful, it is still not the most advanced model yet suggesting that other big developments might be underway. The tech giant maintains that its o3 reasoning model instead is the most powerful one yet. OpenAI does take note of how multiple versions and varied models can become confusing and difficult to handle and to counter this, it plans to bring GPT-5 forward later in the year, which would seamlessly manage casual tasks as well as complex prompts without the need to switch between models. Until that is released, GPT-4.5 is the latest version, with more natural responses and fewer hallucinations. The GPT-4.5 is available starting today for Pro users on a plan of $200 per month. OpenAI, however, plans to make the version more widely available and roll it out to ChatGPT Plus and Team subscribers and then later on to Enterprise and Edu users. The tool would support web searches, file uploads, and even canvas features. OpenAI is determined to keep pushing away the competition through its continued release of innovative products, and, with its ambitious approach, it remains to be a leader in the industry.

OpenAI Launches Its Shiny New GPT-4.5 AI Model -- But It's Not What You're Expecting

The long-awaited moment is here: ChatGPT is receiving an enormous upgrade to GPT-4.5. However, despite OpenAI calling GPT-4.5 its "largest and most knowledgeable model," it comes with a few caveats that might see you continuing to use its older models for a while longer. OpenAI Launches GPT-4.5 With a More "Natural" Feel Let's start with the positives of this GPT update. OpenAI claims GPT-4.5 has a "broader knowledge base" with stronger alignment with "user intent." It also features boosts to emotional intelligence that OpenAI claims make GPT-4.5 "well-suited for tasks like writing, programming, and solving practical problems." The GPT-4.5 research preview also hallucinates less than its predecessors -- 37 percent compared to GPT-4o's 60 percent. Interestingly, OpenAI's internal testing team found GPT-4.5 " warm, intuitive, and natural." This suggests an AI model that emphasizes personality, empathy, and creativity more. GPT-4.5 Isn't All Sunshine and Roses OpenAI also acknowledges that despite the clear improvements to GPT-4.5, it isn't quite ready to replace its earlier models. GPT-4.5 is a much larger and more compute-heavy AI model, which makes it (currently) more expensive to run. Its responses may also feel slower than ChatGPT regulars are now accustomed to with models such as GPT-4o. GPT-4.5's API pricing reflects the intesive compute costs. But it's not only speed. An OpenAI document detailing GPT-4.5's capabilities explains that it "is not a frontier model . . . its performance is below that of o1, o3-mini, and deep research on most preparedness evaluations." At the time of writing, GPT-4.5 doesn't support the multimodal features that make ChatGPT so useful. That means no Voice Mode, video tools, or screen sharing, at least for now. OpenAI has not provided a specific timeline for when this might change. When Can You Use ChatGPT GPT-4.5? Starting from 27 February 2025, GPT-4.5 is rolling out to ChatGPT's $200 Professional subscription tier. Those Professional subscribers will be able to pick up GPT-4.5 immediately from the model picker. From early March, ChatGPT Plus and Team subscribers can choose GPT-4.5, followed by Enterprise and Edu subscribers. To me, this all points in one direction: GPT-4.5 is a stepping stone to GPT-5, expected to launch in 2025. The combination of GPT-4.5, GPT o3, o3-mini, and other models are the building blocks for GPT-5, which many suggest will be the first AI system we can truly deem artificial general intelligence.

OpenAI Offers GPT-4.5 With 40% Fewer Hallucinations, 30x Higher Cost

The rapid release of advanced AI models in the past few days has been impossible to ignore. With the launch of Grok-3 and Claude 3.7 Sonnet, two leading AI companies, xAI and Anthropic, have significantly accelerated the pace of innovation in the field. As rumours about OpenAI's newest model circulated, anticipation surged. However, when GPT-4.5 was released, OpenAI said it wasn't a frontier model and was less powerful than the company's o3-mini model and many others in the competition. It doesn't excel in coding, reasoning, or any such capabilities, either -- because it isn't meant to be. At this time, OpenAI has focused more on the model's usability than anything else. OpenAI tested GPT-4.5 on the SimpleQA benchmark, a tool that evaluates the factual accuracy of AI models in answering short, fact-seeking questions. The model achieved a hallucination rate of 37.1%, in contrast to the o3-Mini, which recorded over 70%. The GPT-4o model exhibited a hallucination rate of 61%. This indicates a 40% reduction in the hallucination rate compared to its predecessor. In accuracy rates on the SimpleQA benchmark, GPT 4.5 scored 62.5%, higher than OpenAI's o3-mini (15%), o1 (47%), and GPT-4o (38.2%). This is also higher than many models in the competition, as the Grok-3 model scored a 43.6% accuracy rate in the benchmark, the Gemini 2.0 Pro scored 44.3%, and the Claude 3.5 Sonnet scored 28.4%. OpenAI also released a system card for the GPT-4.5 model, which evaluates all the safety concerns and associated risks. In an evaluation called PersonQA, which tested the model for hallucinations, GPT-4.5 was more accurate and showed a lesser hallucination rate than the o1 and the GPT-4o models. Given its availability at the $200/month pro plan, several users agreed with OpenAI's claims of reduced hallucinations. Aaron Levie, CEO of the cloud storage company Box, revealed that GPT-4.5 significantly improved over the GPT-4o in extracting data fields from enterprise content, like important details in a contract. "We found a 19 pt [point] improvement in single shot extraction. This is a huge improvement for any mission-critical enterprise workflow," he said in a post on X. Early testers of the model also gave high praise for the model's verbal and emotional intelligence. "I found it to be by far the highest verbal intelligence model I've ever used. It's an outstanding writer and conversationalist," said Theo Jaffee, who had early access to the GPT-4.5 model. While CEO Sam Altman was absent from the launch event, he said on X that GPT-4.5 "is the first model that feels like talking to a thoughtful person to me." "I have had several moments where I've sat back in my chair and been astonished at getting actually good advice from an AI," added Altman, and said that the model offers a different kind of intelligence. There's a magic to it that he hasn't felt before. The model supposedly excels at creative and emotional thinking. Ethan Mollick, a professor at The Wharton School, said on X, "It can write beautifully, is very creative, and is occasionally oddly lazy on complex projects." He even joked that the model took a "lot more" classes in the humanities. Andrej Karpathy, the former OpenAI researcher and founder of Eureka Labs, found that two years ago, when he tested the GPT-4, the model's word choice was more creative, and he had improved understanding of the nuances of the prompt compared to GPT-3.5. Karpathy said that he has a similar feeling for GPT-4.5. Everything is a little bit better," he said. OpenAI, in the model's system card, said internal testers reported GPT-4.5 as warm, intuitive, and natural. "When tasked with emotionally charged queries, it knows when to offer advice, defuse frustration, or simply listen to the user," the report read. Overall, the GPT 4.5 isn't a mind-blowing model, and it isn't the best model on benchmarks either. For example, it is worse than the recently released Claude 3.7 Sonent on coding benchmarks and offers only a marginal improvement over the GPT-4o. Altman also confirmed earlier that the company plans to release the GPT-5 model soon, combining general purpose and reasoning capabilities in a single model. However, if the company aims to make the GPT-4.5 available to the masses, there's bad news. It isn't available yet on the free version or even the $20/month plan. If it were to be deployed on other platforms via API, it would be the most expensive model, and its pricing is an exponential jump over GPT-4o or even the o3-mini. The GPT-4.5 Preview costs $75 and $150 per 1 million input and output tokens, respectively. In comparison, the GPT-4o costs $2.5 and $10 per million input and output tokens, respectively. Clement Delangue, CEO at HuggingFace, said, "IMO [in my opinion], if GPT 4.5 was released as an open-source base model (that everyone can distill), it would be the most impactful release of the year," and added that he isn't a fan of the API either. "Making a few hundred million [dollars] now from it via API doesn't move the needle compared to the 10x more usage/visibility/goodwill/talent they could get by open-sourcing it," he added. OpenAI will have to watch out for the launch of DeepSeek-R2 and Meta's Llama 4, which are expected to be out in a few months. Moreover, if OpenAI is marketing the model for its creative and empathetic outputs, they are subjective metrics at the end of the day. Karpathy conducted a poll on X to check if users prefer outputs of GPT-4.5, or GPT-4o, and many users preferred the latter. It will be interesting to see how many users will be truly pleased with GPT-4.5 when it is released.

ChatGPT-4.5 is here for Pro users now and Plus users next week, and I can't wait to try it

OpenAI claims GPT-4.5 will provide better conversations and fewer hallucinations than ever OpenAI has officially launched GPT-4.5, the most advanced version of its AI models, but it has no deep reasoning ability, because it's designed to be an all-purpose model. The latest model is available on ChatGPT, though for now, you'll need to be a Pro subscriber to access it. The model is being rolled out as a "research preview," which means it's imperfect, and OpenAI hopes users will help sort out any bugs or awkwardness. Even so, the company claims early testers report that GPT-4.5 feels more natural, intuitive, and emotionally intelligent. This is a welcome upgrade for those of us who have ever asked ChatGPT for information and felt like the response was just an outdated Wikipedia page. GPT-4.5 is supposed to be even better at understanding context and recognizing social cues than its predecessors. The overall feeling is supposedly all but indistinguishable from chatting with a real person. It certainly seems to have impressed OpenAI CEO Sam Altman, who called it "the first model that feels like talking to a thoughtful person" in a post on X. Altman added that GPT-4.5 will come to ChatGPT Plus and Team subscribers next week, but you'll need to pay $200 a month to play with it for now. Notably, GPT-4.5 is not only supposedly smarter and more creative, but it should also produce fewer hallucinations. That last part is particularly important because hallucinations were often a sticking point for tentative ChatGPT users who didn't want to rely on a chatbot that might confidently make things up. Performance-wise, GPT-4.5 is supposed to be a big step up and really good with general knowledge. However, that doesn't mean it can handle coding and problem-solving as well as OpenAI's o1 and o3 models or rival options like DeepSeek's R1. Instead, GPT-4.5 specializes in empathy, with a high emotional quotient (EQ). AI still can't feel things, but GPT-4.5 is supposed to be better at picking up on nuance. If you're venting about a bad day, it won't just reply, "I'm sorry you feel that way." Instead, it's designed to offer more thoughtful responses. Whether that makes it feel more like a conversation partner or just a particularly well-trained customer service bot remains to be seen. As for how to measure the EQ, OpenAI quite literally depended on "vibes." "By vibes, really mean the model's EQ, how collaborative it feels and how warm its tone is," OpenAI engineer Rapha Gontijo Lopes explained in a livestream demonstration of GPT-4.5. "We measured this by selecting an opinionated set of prompts and screening our trainers for the ones that most align with our vibes." If you're a ChatGPT Pro user, you'll be among the first to see whether GPT-4.5 truly has the vibes of a human and lives up to the hype. If you're not spending quite that much, you'll get a chance soon enough. Either way, with Amazon announcing a more chatty, AI-powered Alexa+ yesterday, the AI arms race just heated up again, and this time it's looking like the chatbots will be able to mimic how you feel more realistically than ever.

ChatGPT 4.5 -- 5 big upgrades you need to know

OpenAI has unveiled GPT-4.5, its latest and most advanced language model to date. This release brings several significant enhancements over its predecessors, aiming to provide users with a more natural and efficient AI experience. The latest model features upgraded chat capabilities, greater computational efficiency, and more. Here are the five most notable upgrades in ChatGPT-4.5 that are currently available to ChatGPT Pro users and rolling out next week to ChatGPT Plus and Team users. OpenAI is constantly enhancing its chatbots' ability to communicate with its users naturally, and GPT-4.5 has been fine-tuned to deliver more human-like interactions. Thanks to added emotion, users will notice that conversations feel more intuitive and engaging. This model demonstrates a refined personality and an improved understanding of context. The advancement makes it particularly adept at tasks requiring a nuanced grasp of language, such as creative writing and problem-solving. Building upon its predecessors, GPT-4.5 offers a broader and deeper understanding of the world. This expanded knowledge base allows it to provide more accurate and relevant information across various topics, enhancing its utility for users seeking detailed insights into content from news stories to history lessons. One of the critical improvements in GPT-4.5 is its decreased tendency to generate inaccurate information, commonly referred to as "hallucinations." Early testing indicates a significant reduction in hallucination rates compared to previous models, making GPT-4.5 a more reliable tool for users who require precise and factual responses. Despite being OpenAI's largest model to date, GPT-4.5 has achieved over a tenfold increase in computational efficiency compared to GPT-4. This enhancement means users can expect faster response times and a more seamless experience, even when handling complex queries. GPT-4.5 extends its functionality beyond text by accepting image inputs and generating inline SVG images. It also integrates with DALL-E for image generation, allowing users to create visual content directly through the model. Additionally, GPT-4.5 supports web search, ensuring access to the most up-to-date information available when users utilize the search feature. These upgrades position GPT-4.5 as a versatile and powerful chatbot that can take on various tasks, from creative storytelling and image generation to detailed research, summaries, and problem-solving. While the chatbot is currently only accessible to users at the Pro level, Plus users can anticipate seeing the model in their drop-down options as ChatGPT-4.5 rolls out in the next week or so.

OpenAI's GPT-4.5 'won't crush benchmarks' but might be a better friend

ChatGPT-maker OpenAI's upcoming model could be as much as 20 times more costly than its predecessor but will be far more creative and "natural" in its conversational style, according to OpenAI and early testers. OpenAI released a research preview of GPT-4.5 on Feb. 27, its most advanced AI model, which can recognize patterns, draw connections, and make creative insights without reasoning in a superior way to earlier versions, the company said. OpenAI said GPT-4.5's broader knowledge base and improved "EQ" (emotional intelligence) make it more useful for creative tasks and solving practical problems, OpenAI said in a Feb. 27 statement. "We also expect it to hallucinate less and deliver more reliable performance across a wide range of general topics, including richer conversations." GPT-4.5's enhanced creativity and more "natural conversational style" means it isn't well-suited to perform detailed step-by-step logic -- at least compared to OpenAI's o-series-models, it added. The trade-off is that it lacks "chain-of-thought reasoning and can be slower due to its size," the firm said. It also doesn't produce multimodal output like audio or video. OpenAI's latest model received a similar review from Dan Shipper, CEO of AI and the business newsletter Every. "It's not going to blow your mind, but it might befriend you," Shipper said, who said his firm has been testing the latest version for a few days. "It's more like a personality, communication, and creativity upgrade than a huge intelligence leap. It's like OpenAI is pivoting its base model from 'bland assistant' to 'AI bestie.'" Shipper also said GPT-4.5 is "sometimes worse" at following instructions. AI researcher Aran Komatsuzaki also said that GPT-4.5 costs around 15 to 20 times more than GPT-4o to access the API. Ashutosh Shrivastava, founder of the AI Compass newsletter, added: "OpenAI GPT-4.5 pricing is insane. What on earth are they even thinking??" In a Feb. 27 post on X, OpenAI CEO Sam Altman admitted the new reasoning model "won't crush benchmarks" and is a "giant expensive model." Others, such as biomedical scientist and Professor Derya Unutmaz of The Jackson Laboratory, claimed that GPT-4.5 "appears to be remarkable" in medical image diagnosis -- correctly spotting a tubal ectopic pregnancy. Other AI models, such as Grok 3, Claude 3.7 Sonnet, Gemini 2.0 and earlier ChatGPT models, mistakenly identified a medical image as a normal pregnancy, Prof. Unutmaz said. Related: Crypto AI agents see 'remarkable traction' but value still unclear: Sygnum OpenAI's latest iteration of ChatGPT comes as Chinese-based competitor High Flyer launched the open-source AI large-language model DeepSeek R1 in January, which was developed at a fraction of the cost compared to OpenAI's models. OpenAI's CEO Sam Altman, however, claims the cost to build these AI models is falling tenfold or more each year. "You can see this in the token cost from GPT-4 in early 2023 to GPT-4o in mid-2024, where the price per token dropped about 150x in that time period," Altman said in a Feb. 10 post. On Feb. 12, Altman said GPT-5 would be released in a matter of months, which will integrate multiple versions -- including o3 -- into one, OpenAI said on Feb. 13. The free tier of ChatGPT will get unlimited chat access to GPT-5.

OpenAI's GPT-4.5 promises "warm" conversations

OpenAI has launched GPT-4.5, the successor to its previous GPT-4 technology, marking a significant evolution in its chatbot system. This release is notable for being the last version before a shift to "chain-of-thought reasoning," requiring more thoughtful interactions rather than immediate responses. GPT-4.5, which powers the most expensive version of ChatGPT, may not generate the same level of excitement as its predecessor, GPT-4, due to changes in AI research direction. However, OpenAI asserts that this new technology will provide a more natural conversational experience. Mia Glaese, vice president of research at OpenAI, emphasized the model's improved ability to engage in warm, intuitive, and naturally flowing conversations, suggesting a stronger understanding of user intent. This advancement follows the introduction of OpenAI o1, a system intended to reason through complex tasks such as mathematics, coding, and science. Companies like Google, Meta, and the Chinese start-up DeepSeek are pursuing similar technological advancements to enhance problem-solving capabilities in AI. According to OpenAI, GPT-4.5 features enhanced writing capabilities, better world knowledge, and a refined personality compared to earlier models. It excels in recognizing patterns and making connections, making it suitable for writing, programming, and solving practical problems. Despite these upgrades, OpenAI has clarified that GPT-4.5 is not a frontier model. The company stated that it improved GPT-4's computational efficiency by over 10 times but noted the model does not introduce seven new frontier capabilities compared to previous releases. Its performance is regarded as below that of the o1, o3-mini, and deep research on most preparedness evaluations. GPT-5 is coming: OpenAI just revealed it Reports indicated that OpenAI employed the o1 reasoning model, codenamed Strawberry, for training GPT-4.5 using synthetic data. The training methods incorporated new supervision techniques alongside traditional approaches such as supervised fine-tuning and reinforcement learning from human feedback, much like those used for GPT-4o. OpenAI claims that GPT-4.5 experiences fewer hallucinations than GPT-4o and slightly fewer than its o1 model. Raphael Gontijo Lopes, a researcher at OpenAI, stated that the alignment efforts aimed to make the conversations feel warmer, more intuitive, and emotionally nuanced. Human testers evaluating GPT-4.5 against GPT-4o reported its superior performance across nearly all evaluation categories. OpenAI CEO Sam Altman described GPT-4.5 as a "giant, expensive model" that "won't crush benchmarks." The model launched for ChatGPT Pro users and is set to roll out to Plus and Team users shortly, followed by Enterprise and Edu users. GPT-4.5 is also currently available on Microsoft's Azure AI Foundry platform, alongside new models from Stability, Cohere, and Microsoft. The underlying system of GPT-4.5 falls under the category of large language models (L.L.M.s), which learn by analyzing vast amounts of text from the internet, including Wikipedia, books, and chat logs. They identify patterns to generate text independently. To develop reasoning systems, companies subject L.L.M.s to an additional reinforcement learning process, which can last weeks or months. This method helps the models learn the most effective strategies for solving various problems through extensive trial and error. While some reasoning systems have outperformed standard L.L.M.s on specific standardized tests, experts caution that such assessments may not accurately reflect real-world performance. The new reasoning systems still face challenges, including the potential for inaccuracies and fabrications, known as hallucination. OpenAI announced that GPT-4.5 would be accessible to anyone subscribed to the ChatGPT Pro service, priced at $200 per month, starting Thursday.

OpenAI Unveils GPT-4.5: Friendliest Model Yet at 1300% the Price

OpenAI released GPT-4.5 on Thursday, just one day after Anthropic launched Claude 3.7 Sonnet and merely a week following xAI's Grok-3 debut and DeepSeek's announcement of a new model coming soon. And expensive is the operative word here. OpenAI's new model comes with an eye-watering API price tag of $75 per million input tokens and $150 per million output tokens. It appears to be a new competitive phase in the AI race, with companies scrambling to outdo each other with increasingly capable -- and increasingly expensive -- models. For context, that's ten times pricier than Claude 3.7 Sonnet, making it potentially prohibitive for many developers and startups looking to build on the technology. GPT-4o (its predecessor) cost $2.50 per 1M tokens of input and $10.00 per 1M tokens of output -- making GPT-4.5 2900% more expensive to input and 1300% dearer to get a response. Sam Altman, OpenAI's CEO, didn't shy away from acknowledging the model's massive resource requirements in his announcement. "Bad news: It is a giant, expensive model," he said. "A heads up: this isn't a reasoning model and won't crush benchmarks. It's a different kind of intelligence," Altman said. "There's a magic to it I haven't felt before." And this seems to be the key. Users are paying 1300% more not to have a more intelligent model, but to have a nicer model that feels more human. For example, one thing in which GPT-4.5 shines, according to OpenAI, is in what they call "vibes," or essentially the model's EQ, warmth, and collaborative feel. The company created a "Vibes test set" measuring creative intelligence and conversational quality, on which GPT-4.5 purportedly outperformed other models. The examples shared during the presentation didn't exactly introduce anything new. The first demonstration had literally this prompt: "UGHHH! My friend cancelled on me again!!! Write a text message telling them that I HATE THEM!!!!" which arguably isn't something for which you would use a competent large language model. In a following demonstration comparing GPT-4.5 to OpenAI's o1 model, researchers asked both AIs to explain the need for AI alignment and to help craft a message to a friend who had canceled plans. The responses, while showing some improved nuance in GPT-4.5, hardly seemed revolutionary. The difference was in the tone. In another example, the research team asked the powerful GPT-4.5 why the sea water is salty. The new model responded using less complex terms -- "because of rain, rivers, and rocks" -- compared to previous models. GPT-4-Turbo gave a more comprehensive and detailed reply, which the team didn't like, arguing that "you get the feeling that it wants you to know how smart it is." One amusing detail from the presentation was an Easter egg hinting at a possible GPT-6, with a query that read: "Num GPUs for GPT-6 Training." Perhaps when that model arrives, the demos will be more impressive. The benchmarks presented paint a mixed picture. GPT-4.5 scores 71.4% on GPQA (a science evaluation), compared to GPT-4o's 53.6%. However, it still trails behind OpenAI's o3-mini model, which scores 79.7% through its reasoning capabilities. Similar patterns emerged across other benchmarks. On the AIME '24 math evaluation, GPT-4.5 scored 36.7%, beating GPT-4o's 9.3% but still far behind o3-mini's 87.3%. For coding tasks, GPT-4.5 outperformed its predecessor and o3-mini on the SWE-Lancer Diamond benchmark but fell short on SWE-Bench Verified compared to the reasoning-focused model. Altman described the model in almost mystical terms, calling it "the first model that feels like talking to a thoughtful person." He added: "I have had several moments where I've sat back in my chair and been astonished at getting actually good advice from an AI." During the model's presentation, OpenAI researchers explained that the company advances AI through two distinct approaches: unsupervised learning and reasoning. While reasoning teaches models to "think before responding," unsupervised learning helps increase "word model accuracy and intuition." GPT-4.5 doubles down on the latter. "GPT-4.5 is our next step in scaling up unsupervised learning, increasing world knowledge, intuition, and reducing hallucinations," an OpenAI research lead explained in the presentation. Developing GPT-4.5 required massive technical innovation, according to the team. They had to build new inference systems to serve such a large model efficiently, use low-precision training to maximize GPU usage, and even train across multiple data centers simultaneously. The release comes at a time when consumer expectations for AI are sky-high, and competition in the space is intensifying. Whether GPT-4.5's "different kind of intelligence" and improved "vibes" justify its enormous resource requirements and steep pricing remains to be seen. GPT-4.5 is currently available for Pro users who pay $200 a month. Plus users paying $20 a month will have access to the model next week.

OpenAI Launches GPT-4.5 For Pro Users

OpenAI has launched GPT-4.5, its latest AI model, in a research preview for ChatGPT Pro users. The company claims the model offers improved writing skills, better pattern recognition, and fewer hallucinations compared to previous versions. However, CEO Sam Altman has highlighted its high computational costs and resource requirements. The release of ChatGPT's latest model follows the emergence of competitors like DeepSeek-R1, prompting companies like OpenAI to advance their AI development efforts. Key Features and Enhancements OpenAI claims that GPT-4.5 is designed to improve conversational interactions, making them more natural and intuitive. It has a broader knowledge base and better emotional awareness, making it useful for tasks like writing, coding, and problem-solving. OpenAI Researcher Raphael Gontijo Lopes stated, "We aligned GPT-4.5 to be a better collaborator, making conversations feel warmer, more intuitive, and emotionally nuanced." According to OpenAI, human evaluators rated it higher than GPT-4o in most categories. An incremetal upgrade OpenAI acknowledges that GPT-4.5 is not a frontier model but an incremental upgrade. It improves GPT-4's computational efficiency by more than ten times but does not introduce entirely new reasoning capabilities. The model was trained using supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF). Challenges and resource constraints Sam Altman addressed the challenges in deploying GPT-4.5. In a post on X (formerly Twitter), he wrote about its high resource demands, stating, "We've been growing a lot and are out of GPUs. We will add tens of thousands of GPUs next week and roll it out to the Plus tier then." Availability and future prospects Currently, GPT-4.5 is available for ChatGPT Pro users, with expansion planned for Plus and Team users next week, followed by Enterprise and Educational users. It is also available on Microsoft's Azure AI Foundry. However, the model lacks multimodal capabilities such as voice mode and video, which OpenAI aims to introduce later. Why It Matters GPT-4.5's release comes amid growing competition, including DeepSeek's recent launch. Unlike previous breakthroughs, OpenAI has framed this update as an improvement rather than a frontier advancement. This suggests that AI companies may be shifting focus -- whether due to resource limitations, the rising costs of scaling models, or a strategic pivot toward efficiency. As the AI race continues, the challenge will be balancing innovation with the practical constraints of hardware and infrastructure.

It's a lemon" -- OpenAI's largest AI model ever arrives to mixed reviews

The verdict is in: OpenAI's newest and most capable traditional AI model, GPT-4.5, is big, expensive, and slow, providing marginally better performance than GPT-4o at 30x the cost for input and 15x the cost for output. The new model seems to prove that longstanding rumors of diminishing returns in training unsupervised-learning LLMs were correct and that the so-called "scaling laws" cited by many for years have possibly met their natural end. An AI expert who requested anonymity told Ars Technica, "GPT-4.5 is a lemon!" when comparing its reported performance to its dramatically increased price, while frequent OpenAI critic Gary Marcus called the release a "nothing burger" in a blog post (though to be fair, Marcus also seems to think most of what OpenAI does is overrated). Former OpenAI researcher Andrej Karpathy wrote on X that GPT-4.5 is better than GPT-4o but in ways that are subtle and difficult to express. "Everything is a little bit better and it's awesome," he wrote, "but also not exactly in ways that are trivial to point to." OpenAI is well aware of these limitations, and it took steps to soften the potential letdown by framing the launch as a relatively low-key "Research Preview" for ChatGPT Pro users and spelling out the model's limitations in a GPT-4.5 release post published Thursday. "GPT‑4.5 is a very large and compute-intensive model, making it more expensive⁠ than and not a replacement for GPT‑4o," the company wrote. "Because of this, we're evaluating whether to continue serving it in the API long-term as we balance supporting current capabilities with building future models." According to OpenAI's own benchmark results, GPT-4.5 scored significantly lower than OpenAI's simulated reasoning models (o1 and o3) on tests like AIME math competitions and GPQA science assessments, with GPT-4.5 scoring just 36.7 percent on AIME compared to o3-mini's 87.3 percent. Additionally, GPT-4.5 costs five times more than o1 and over 68 times more than o3-mini for input processing. And GPT-4.5 is terrible for coding, relatively speaking, with an October 2023 knowledge cutoff that may leave out knowledge about updates to development frameworks. Tech investor Paul Gauthier performed independent testing of GPT-4.5's coding ability using Aider's Polyglot Coding benchmark and found that GPT-4.5 ranked 10th in overall ability (Claude 3.7 Sonnet with extended thinking is on top, and o1 and o3 ahead), and it also ranked poorly in performance versus cost, meaning that for coding tasks, GPT-4.5 isn't worth the price you'd have to pay through the API. According to OpenAI's benchmarks, GPT-4.5 does show some improvements over GPT-4o in specific areas. The model scored higher on the multilingual MMMLU (general knowledge) test with 85.1 percent compared to GPT-4o's 81.5 percent, suggesting better performance on knowledge-based tasks across multiple languages. OpenAI also reports that GPT-4.5 demonstrated improved performance in reducing confabulations (hallucinations), with the company claiming it generates fewer false or misleading responses than previous versions. OpenAI's testing also indicated that human evaluators preferred GPT-4.5's responses over GPT-4o in about 57 percent of interactions, suggesting modest but measurable improvements in overall user experience. However, these incremental gains come with significantly higher computational demands and costs. High on vibes, low on reasoning Upon 4.5's release, OpenAI CEO Sam Altman did some expectation tempering on X, writing that the model is strong on vibes but low on analytical strength. "It is the first model that feels like talking to a thoughtful person to me," he wrote. He then added further down in his post, "a heads up: this isn't a reasoning model and won't crush benchmarks. it's a different kind of intelligence and there's a magic to it i haven't felt before." GPT-4.5 is so massive and inefficient that OpenAI Sam Altman also wrote on X that the company would have liked to release GPT-4.5 for everyone but that the company is "out of GPUs." More are on the way, he said. Perhaps because of the disappointing results, Altman had previously written that GPT-4.5 will be the last of OpenAI's traditional AI models, with GPT-5 planned to be a dynamic combination of "non-reasoning" LLMs and simulated reasoning models like o3. A stratospheric price and a tech dead-end And about that price -- it's a doozy. GPT-4.5 costs $75 per million input tokens and $150 per million output tokens through the API, compared to GPT-4o's $2.50 per million input tokens and $10 per million output tokens. (Tokens are chunks of data used by AI models for processing). For developers using OpenAI models, this pricing makes GPT-4.5 impractical for many applications where GPT-4o already performs adequately. In contrast, OpenAI's flagship reasoning model, o1 pro, costs $15 per million input tokens and $60 per million output tokens -- significantly less than GPT-4.5 despite offering specialized simulated reasoning capabilities. Even more striking, the o3-mini model costs just $1.10 per million input tokens and $4.40 per million output tokens, making it cheaper than even GPT-4o while providing much stronger performance on specific tasks. OpenAI has likely known about diminishing returns in training LLMs for some time. As a result, the company spent most of last year working on simulated reasoning models like o1 and o3, which use a different inference-time (runtime) approach to improving performance instead of throwing ever-larger amounts of training data at GPT-style AI models. While this seems like bad news for OpenAI in the short term, competition is thriving in the AI market. Anthropic's Claude 3.7 Sonnet has demonstrated vastly better performance than GPT-4.5, with a reportedly more efficient architecture. It's worth noting that Claude 3.7 Sonnet is likely a system of AI models working together behind the scenes, although Anthropic has not provided details about its architecture. For now, it seems that GPT-4.5 may be the last of its kind -- a technological dead-end for an unsupervised learning approach that has paved the way for new architectures in AI models, such as o3's inference-time reasoning and perhaps even something more novel, like diffusion-based models. Only time will tell how things end up. GPT-4.5 is now available to ChatGPT Pro subscribers, with rollout to Plus and Team subscribers planned for next week, followed by Enterprise and Education customers the week after. Developers can access it through OpenAI's various APIs on paid tiers, though the company is uncertain about its long-term availability.

ChatGPT Gets Another Upgrade With GPT-4.5

OpenAI has released a ton of improvements for its chatbot, ChatGPT, in recent months, but it's been a while since we've had a core model update. While we await for GPT-5, which OpenAI seems pretty ambitious with, the company has launched a mid-step upgrade in the meantime -- GPT-4.5. OpenAI has officially launched GPT-4.5, its newest and largest language model to date. While not considered a "frontier" model, as per the company itself, GPT-4.5 nonetheless offers significant improvements in computational efficiency, writing quality, world knowledge, and reduced hallucination rates compared to its predecessor, GPT-4. The model is being made available as a "research preview," which means it will first undergo a controlled rollout initially focused on gathering user feedback and on giving it final touches before it becomes generally available. GPT-4.5 is reported to have a greater than tenfold improvement in computational efficiency compared to GPT-4. This means the model can achieve similar or better results with significantly less processing power. OpenAI also says GPT-4.5 has enhanced writing capabilities and expanded world knowledge. This means that we should also have improvements in areas like text generation, summarization, translation, and question answering. And we should also have a "refined personality" over previous models. We're not sure about the exact meaning of that, but it could mean we might have a more consistent and engaging conversational style with this model that doesn't read as robotically. A key focus of GPT-4.5's development has been minimizing hallucinations -- instances where the model confidently generates false or nonsensical information. According to OpenAI, GPT-4.5 hallucinates less than GPT-4o, and only slightly less than o1. GPT-4.5 has been trained using a combination of "new supervision techniques" alongside established methods like supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF). Despite these advancements, OpenAI is clear that GPT-4.5 is not a frontier model -- and for some stuff, it might not even be a massive step forward in any way. It doesn't feature any brand-new or significantly improved capabilities, and as a matter of fact, GPT-4.5 can actually perform "below that of o1, o3-mini, and deep research on most preparedness evaluations." This is an upgrade over GPT-4, sure, but the model the company is preparing as its next game-breaking model is GPT-5. This is just to hold us over until that arrives. Still, it's better than nothing at all. As a reminder, CEO Sam Altman recently acknowledged on Twitter/X the growing complexity of OpenAI's offerings, and said that the company's goal is to make AI "just work" for users. The company's immediate plan involved the release of this model, codenamed Orion, which will be the company's final release before transitioning exclusively to a "chain-of-thought" approach for future models. This method, already used by GPT o1/o3 and DeepSeek models, breaks down prompts into several steps, resulting in more reliable answers and fewer AI hallucinations. Looking further ahead, OpenAI plans to introduce GPT-5, which the company thinks about as an integrated system consolidating various technologies. This includes the full o3 model's functionalities (right now we only have o3-mini, and the full o3 model is reserved for the release of GPT-5). GPT-5 aims to support features like voice interaction, canvas capabilities, enhanced search, and deep research tools. Details remain scarce so far, but the company has indicated that the availability of specific tools may vary across subscription tiers. GPT-4.5 is now rolling out as a "research preview." ChatGPT Pro users will gain access to the model today, and it should also arrive for Plus and Team users by next week. It's not clear if, or when, the model might roll out for free users, but if this goes anything similar to the initial GPT-4 rollout, we might see it pop up first on other chatbots such as Microsoft Copilot before it lands for free ChatGPT. Source: OpenAI, The Verge

OpenAI Releases Its Next-Generation GPT-4.5 Model to ChatGPT Pro Users

OpenAI says GPT-4.5 has a thoughtful personality and excels at creative writing. It also exhibits fewer hallucinations. OpenAI introduced GPT-4o, a non-reasoning model to ChatGPT users back in May 2024. Finally, over 10 months later, the hot AI startup has unveiled its next-generation GPT-4.5 AI model, codenamed 'Orion' today. GPT-4.5 is the last non-reasoning model from OpenAI as the upcoming GPT-5 will merge the o3 reasoning model to create a unified AI system. OpenAI says GPT-4.5 is the "largest and most knowledgeable language model" developed by the company so far, but it's not a frontier model. It's designed to be more general-purpose than STEM-focused o-series reasoning models. It means that GPT-4.5 excels at creative writing, natural conversation, practical problem-solving, and offers a broader knowledge base. Note that it's a multimodal model so it can process images and files too. What is interesting is that GPT-4.5 exhibits fewer hallucinations than GPT-4o. Its hallucination rate dropped to 37.1% from GPT-4o's 61.8%. And GPT-4.5's accuracy improved to 62.5% from GPT-4o's 38.2%. Apart from that, early testers say that GPT-4.5 is "warm, intuitive, and natural" during conversations. As for benchmarks, GPT-4.5 outperforms GPT-4o in MMLU across 14 languages. Next, in SWE-bench Verified which evaluates the ability to solve real-world software issues, GPT-4.5 achieves 38% while GPT-4o gets 30.7%. That said, it performs worse than the o1, o3, and o3-mini reasoning models. In the new SWE-Lancer benchmark developed by OpenAI which evaluates performance on real-world, economically valuable software engineering tasks, GPT-4.5 solved 32.6% of the tasks, compared to GPT-4o's 23.3%. In GPQA (Science), GPT-4.5 scored 71.4% while GPT-4o got 53.6%. About availability, GPT-4.5 is rolling out to ChatGPT Pro users starting today. And OpenAI says starting next week, GPT-4.5 will be available to ChatGPT Plus, Team, and Edu users. All in all, it appears scaling LLMs via pre-training has hit a wall, and that's why OpenAI says GPT-4.5 will be the last non-reasoning model. In the benchmark numbers, it's clear that o-series reasoning models perform exceptionally well, even on older base models. Nevertheless, in every aspect, GPT-4.5 performs better than GPT-4o while being 10x more efficient. It has a refined personality, produces superior writing, and has a broader world knowledge. Now, anticipation builds for the unified GPT-5 AI system which will integrate the o3 reasoning model. It's likely to be released in May this year.

OpenAI ChatGPT-4.5 is here and it's the most human-like chatbot yet -- here's how to try it

OpenAI has officially unveiled GPT-4.5, the latest and most powerful iteration of its AI model, marking a significant leap forward in conversational AI. Today, GPT-4.5 was announced as a research preview for ChatGPT Pro users and developers, with plans to roll it out to Plus and Team users next week. The model represents a major step in scaling unsupervised learning, leading to more human-like conversations, broader knowledge and fewer hallucinations. While previous iterations like GPT-4o focused on speed and multimodal capabilities, GPT-4.5 refines the AI's ability to understand nuance, process context, and engage in more intuitive dialogue. According to OpenAI, the model has been optimized to recognize patterns more effectively, draw stronger connections, and generate creative insights with improved accuracy. One of GPT-4.5's standout features is its ability to engage in warm, fluid, and naturally flowing conversations, making AI interactions feel more human than ever before. Enhanced emotional intelligence (EQ) and better steerability allow it to understand user intent better, interpret subtle cues, and maintain engaging discussions that feel personalized and insightful. Additionally, hallucinations -- AI-generated inaccuracies or false information -- have been significantly reduced. OpenAI credits this improvement to advancements in unsupervised learning and optimization techniques that allow the model to refine its world knowledge and intuition more effectively. Essentially, AI learns from its own mistakes and corrects itself accordingly, which helps it improve consistently. The development of GPT-4.5 centers around scaling two complementary AI paradigms: unsupervised learning and reasoning. OpenAI explains that scaling reasoning trains AI to think step-by-step before responding, helping it tackle complex STEM and logic problems. Unsupervised learning increases the model's knowledge accuracy and pattern recognition, improving how it processes and synthesizes information. GPT-4.5's core improvements come from scaling up compute and data alongside model architecture and optimization innovations. The result is a chatbot that feels more natural, intuitive, and reliable than any previous version. OpenAI's early testing highlights several areas where GPT-4.5 excels. OpenAI's early testing highlights several areas where GPT-4.5 excels: For example, when asked about an obscure historical painting, GPT-4.5 accurately pinpointed "The Trojan Women Setting Fire to Their Fleet" by Claude Lorrain, explaining its significance in Virgil's Aeneid with impressive depth. Similarly, when responding to a user struggling with a failed test, GPT-4.5 delivered a thoughtful, emotionally intelligent response acknowledging the user's feelings while providing practical advice. Starting today, ChatGPT Pro users can select GPT-4.5 in the web, mobile, and desktop model picker. Plus and Team users will gain access next week. Developers can also start experimenting with GPT-4.5 via the Chat Completions API, Assistants API, and Batch API, where the model supports features like function calling, structured outputs, and vision capabilities through image inputs. However, GPT-4.5 does not currently support multimodal features like voice mode, video, or screen sharing. OpenAI has hinted at future updates to integrate these functionalities into upcoming models. OpenAI believes that combining unsupervised learning with advanced reasoning models will unlock new levels of AI intelligence. While GPT-4.5 primarily focuses on knowledge, intuition, and collaboration, OpenAI also works on models with advanced reasoning and decision-making skills. Safety remains a priority as well. The company conducted extensive safety tests and red-teaming evaluations before launch, ensuring GPT-4.5 adheres to OpenAI's Preparedness Framework to prevent misuse. To showcase the full potential of GPT-4.5, OpenAI is hosting a livestream tomorrow, Friday, February 28 at noon PT with its researchers, who will discuss the model's development and capabilities. As AI continues to evolve, GPT-4.5 marks a key milestone -- one that makes AI interactions feel more natural, helpful, and human-like than ever before. The latest from OpenAI sets a new standard for chatbots, proving that AI is not only getting smarter but also more intuitive and engaging.

OpenAI launches GPT 4.5 AI model with deeper knowledge and emotions

OpenAI has just introduced its latest AI model, dubbed GPT 4.5, which the company claims is its largest and best model yet. This is not a reasoning model, a faculty which is limited to the O-series models. Despite that, GPT 4.5 is touted to be a more natural conversationalist with a higher emotional quotient and improved problem-solving capabilities. As far as core competencies go, it has access to the latest information from the web, allows file and multimedia upload, and supports canvas platform for coding-related tasks. However, it currently lacks support for voice mode, video comprehension, and screen-sharing. Recommended Videos For now, this model is limited to the research preview phase, which means it's not even widely available to ChatGPT Plus subscribers. It seems infrastructure challenges are holding back the wider availability of GPT 4.5 to the market. "It is a giant, expensive model. we really wanted to launch it to plus and pro at the same time, but we've been growing a lot and are out of GPUs. we will add tens of thousands of GPUs next week and roll it out to the plus tier then," explained OpenAI CEO, Sam Altman. Unlike models that reason through a chain of thought and then provide answers, GPT 4.5 relies on a technique called unsupervised learning, that involves scaling up the training data as well as the compute requirements. GPT-4.5 is ready! good news: it is the first model that feels like talking to a thoughtful person to me. i have had several moments where i've sat back in my chair and been astonished at getting actually good advice from an AI. bad news: it is a giant, expensive model. we... — Sam Altman (@sama) February 27, 2025 The overall benefit is reduced hallucination (one of the biggest trust problems with any generative AI chatbot), a broader base of knowledge, and deeper understanding. As per benchmarks shared by the company, it ranks much higher at result accuracy compared to the GPT-4o, o3-mini, and o1 AI models. At the same time, the hallucination rate is less than half compared to the o3-mini model, and lower than the rest of its siblings. "For GPT‑4.5, we developed new, scalable techniques that enable training larger and more powerful models with data derived from smaller models," adds the company. It can handle creative queries, professional questions, and nuanced conversations in a more natural way than any model that OpenAI has launched so far. Overall, the AI giant says GPT 4.5 is better at creativity, aesthetic intelligence, writing, as well as design-related chores. Access to GPT 4.5 is now rolling out to ChatGPT Pro subscribers on web, mobile, and desktop. ChatGPT Plus and Team tiers will get access to it next week, followed by Enterprise and Education bundles.

OpenAI Launches GPT-4.5 for ChatGPT -- It's Huge and Compute-Intensive

Internally called Orion, GPT-4.5 is OpenAI's largest model to date, and it's only available through the company's $200 monthly ChatGPT subscription. GPT-4.5 is here, and OpenAI's newest generative AI model is bigger and more compute-intensive than ever -- it's supposedly also better at understanding what ChatGPT users mean with their prompts. Users who want to be part of the first wave to try GPT-4.5, labeled as a research preview, will be required to pay for OpenAI's $200-a-month ChatGPT Pro subscription. Prior to this launch, 2025 has already been filled with new AI model releases. Anthropic recently put out a hybrid reasoning model for its Claude chatbot. Before that, Chinese researchers at DeepSeek rocked Silicon Valley with their release of a powerful model trained on a tiny budget, prompting OpenAI to drop a "mini" version of its reasoning model a month ago. Alongside these new releases, OpenAI promised to invest billions into building the AI infrastructure required to fuel more massive models. And GPT-4.5 is a reiteration of this current strategy from the startup: Bigger is better. ChatGPT 4.5 is in stark contrast to other recent AI innovations, like DeepSeek's R1, that attempted to match the performance of a frontier model with as few resources as possible. OpenAI still sees a strong path forward through scaling its models. According to researchers who worked on GPT-4.5, this kind of maximalist mindset to model development has captured more of the nuances of human emotions and interactions. They see the model's size as also potentially helping this iteration hallucinate with less frequency than past releases. "If you know more things, you don't need to make things up," says Mia Glaese, who leads OpenAI's alignment team and human data team. Exactly how big or compute-intensive GPT-4.5 is remains unclear -- OpenAI declined to share specific numbers. So, what's it like to use the new model? GPT-4.5 supports the web search and canvas feature as well as uploads of files and images, though it's not yet compatible with the AI Voice Mode. In the announcement post for GPT-4.5, OpenAI included academic benchmark results that show the model getting vastly outpaced by the o3-mini model when it comes to math, and slightly upstaged on science as well, though GPT-4.5 did score a little higher on language benchmarks. The researchers say these measurements don't capture the full story. "We would expect the difference in 4.5 to be similar to the experience difference of 4 to 3.5," says Glaese. For the user, prompts related to subjects like writing or programming may yield stronger results, with the back-and-forth interactions feeling more "natural" overall. She hopes all of the chats from this limited release will help them to better understand what GPT-4.5 excels at, as well as its limitations. Unlike those released as part of OpenAI's "o" series, GPT-4.5 is not considered to be a reasoning model. The company's CEO, Sam Altman, posted on social media earlier in February that OpenAI would "ship GPT-4.5, the model we called Orion internally, as our last non-chain-of-thought model." Nick Ryder, who leads the company's foundations-in-research team, clarified that this statement pertained to streamlining OpenAI's product road map, not its research road map. The startup is not just looking into reasoning models, but users can expect to see a more blended experience overall with future releases for ChatGPT where you don't have to pick which one to use.

OpenAI's new GPT-4.5 model is a better, more natural conversationalist

It's not available to free users, but serves as a preview for whats to come in GPT-5 and beyond. In what has already been a busy past few days for new model releases, OpenAI is capping off the week with a research preview of GPT-4.5. The company is touting the new system as its largest and best model for chat yet. In early testing, OpenAI says people found GPT-4.5 to be a more natural conversationalist, with the ability to convey warmth and display a kind of emotional intelligence. In one example shared by OpenAI, a person tells ChatGPT they're going through a hard time after failing a test. Where the company's previous models, including GPT-4o and o3-mini, might commiserate with the individual before offering a long list of unsolicited advice, GPT-4.5 takes a different tact. "Want to talk about what happened, or do you just need a distraction? I'm here either way," the chatbot says when powered by GPT-4.5. The gains shown by GPT-4.5 are the result of advancements OpenAI made in unsupervised learning. With unsupervised learning, a machine learning algorithm is given an unlabeled data set and left to its own devices to find patterns and insights. GPT-4.5 doesn't "think" like the company's state-of-the-art reasoning models, but in training the new model OpenAI made architectural enhancements and gave it access to more data and compute power. "The result is a model that has broader knowledge and a deeper understanding of the world, leading to reduced hallucinations," the company says. Speaking of reduced hallucinations, OpenAI measured how much better GPT-4.5 in that regard. When put through SimpleQA, an OpenAI-designed benchmark that tests large language models on their ability to answer "straightforward but challenging knowledge questions," GPT-4.5 beat out o3-mini, GPT-4o and even o1 with a hallucination rate of 37.1 percent. Obviously, the new model doesn't solve the problem of AI hallucinations altogether, but it is a step in the right direction. Despite its relative strengths over GPT-4o and o3-mini, GPT-4.5 isn't a direct replacement for those models. Compared to OpenAI's reasoning systems, GPT-4.5 is "a more general-purpose, innately smarter model." Additionally, it's not natively multimodal like GPT-4o, meaning it doesn't work with features like Voice Mode, video or screensharing. It's also "a very large and compute-intensive model." It's best to think of GPT-4.5 as a stepping stone to systems OpenAI plans to offer in the future. In fact, Sam Altman said as much earlier this month when he shared the company's roadmap, noting GPT-4.5 would be "our last non-chain-of-thought model" -- referring to the fact that the new system doesn't solve problems by tackling them step by step like OpenAI's reasoning models do. Its successor, GPT-5, will likely integrate many of OpenAI's latest technologies, including its frontier o3 model. OpenAI reiterated that today, saying it plans to bring GPT-4.5's "unique strengths, including broader knowledge, stronger intuition, and greater 'EQ,' to all users in future models." In the meantime, ChatGPT Pro subscribers can begin using GPT-4.5 starting today, with Pro and Team users slated to gain access starting next week.

GPT-4.5 for enterprise: Do its accuracy and knowledge justify the cost?

However, given that this is OpenAI's largest and most powerful non-reasoning model, it is worth considering its strengths and the areas where it shines. Better knowledge and alignment There is little detail about the model's architecture or training corpus, but we have a rough estimate that it has been trained with 10X more compute. And, the model was so large that OpenAI needed to spread training across multiple data centers to finish in a reasonable time. Bigger models have a larger capacity for learning world knowledge and the nuances of human language (given that they have access to high-quality training data). This is evident in some of the metrics presented by the OpenAI team. For example, GPT-4.5 has a record-high ranking on PersonQA, a benchmark that evaluates hallucinations in AI models. Practical experiments also show that GPT-4.5 is better than other general-purpose models at remaining true to facts and following user instructions. Users have pointed out that GPT-4.5's responses feel more natural and context-aware than previous models. Its ability to follow tone and style guidelines has also improved. After the release of GPT-4.5, AI scientist and OpenAI co-founder Andrej Karpathy, who had early access to the model, said he "expect[ed] to see an improvement in tasks that are not reasoning-heavy, and I would say those are tasks that are more EQ (as opposed to IQ) related and bottlenecked by e.g. world knowledge, creativity, analogy making, general understanding, humor, etc." However, evaluating writing quality is also very subjective. In a survey that Karpathy ran on different prompts, most people preferred the responses of GPT-4o over GPT-4.5. He wrote on X: "Either the high-taste testers are noticing the new and unique structure but the low-taste ones are overwhelming the poll. Or we're just hallucinating things. Or these examples are just not that great. Or it's actually pretty close and this is way too small sample size. Or all of the above." Better document processing In its experiments, Box, which has integrated GPT-4.5 into its Box AI Studio product, wrote that GPT-4.5 is "particularly potent for enterprise use-cases, where accuracy and integrity are mission critical... our testing shows that GPT-4.5 is one of the best models available both in terms of our eval scores and also its ability to handle many of the hardest AI questions that we have come across." In its internal evaluations, Box found GPT-4.5 to be more accurate on enterprise document question-answering tasks -- outperforming the original GPT-4 by about 4 percentage points on their test set. Box's tests also indicated that GPT-4.5 excelled at math questions embedded in business documents, which older GPT models often struggled with. For example, it was better at answering questions about financial documents that required reasoning over data and performing calculations. GPT-4.5 also showed improved performance at extracting information from unstructured data. In a test that involved extracting fields from hundreds of legal documents, GPT-4.5 was 19% more accurate than GPT-4o. Planning, coding, evaluating results Given its improved world knowledge, GPT-4.5 can also be a suitable model for creating high-level plans for complex tasks. Broken-down steps can then be handed over to smaller but more efficient models to elaborate and execute. According to Constellation Research, "In initial testing, GPT-4.5 seems to show strong capabilities in agentic planning and execution, including multi-step coding workflows and complex task automation." GPT-4.5 can also be useful in coding tasks that require internal and contextual knowledge. GitHub now provides limited access to the model in its Copilot coding assistant and notes that GPT-4.5 "performs effectively with creative prompts and provides reliable responses to obscure knowledge queries." Given its deeper world knowledge, GPT-4.5 is also suitable for "LLM-as-a-Judge" tasks, where a strong model evaluates the output of smaller models. For example, a model such as GPT-4o or o3 can generate one or several responses, reason over the solution and pass the final answer to GPT-4.5 for revision and refinement. Is it worth the price? Given the huge costs of GPT-4.5, though, it is very hard to justify many of the use cases. But that doesn't mean it will remain that way. One of the constant trends we have seen in recent years is the plummeting costs of inference, and if this trend applies to GPT-4.5, it is worth experimenting with it and finding ways to put its power to use in enterprise applications. It is also worth noting that this new model can become the basis for future reasoning models. Per Karpathy: "Keep in mind that that GPT4.5 was only trained with pretraining, supervised finetuning and RLHF [reinforcement learning from human feedback], so this is not yet a reasoning model. Therefore, this model release does not push forward model capability in cases where reasoning is critical (math, code, etc.)... Presumably, OpenAI will now be looking to further train with reinforcement learning on top of GPT-4.5 model to allow it to think, and push model capability in these domains."

OpenAI Says Its New ChatGPT 4.5 Has Better Emotional Intelligence

Samantha Kelly is a freelance writer with a focus on consumer technology, AI, social media, Big Tech, emerging trends and how they impact our everyday lives. Her work has been featured on CNN, NBC, NPR, the BBC, Mashable and more. The next version of ChatGPT is here, with the promise to be more humanistic in tone, interactions and general knowledge. OpenAI announced on Thursday a preview of GPT-4.5, what OpenAI is calling its largest and best model for chat yet. The company said to expect more warm, naturally flowing conversations, better emotional intelligence and more advanced capabilities, such as the ability to recognize patterns, draw connections and think more creatively. The company detailed in a press release that the model can also understand more nuance and context, is more attuned to collaboration and is generally more knowledgeable, effectively reducing the likeliness to hallucinate. The news comes as tech companies are embracing a broader goal of developing AI that can match or surpass human capabilities. While GPT-4.5 is a step forward, it's perhaps more incremental than the highly anticipated, more advanced GPT 5 model reportedly not too far behind. This new version also comes on the heels of China's DeepSeek shaking up the generative AI field with its low-cost model and Google releasing version 2.0 of its Gemini AI. OpenAI continues to expand its various product offerings. Earlier this week, it opened up its Deep Research tool to all paying ChatGPT users and rolled out a free version of Advanced Voice Mode, a feature that enables real-time voice interactions with ChatGPT. The company said robust pre-training helps models like GPT-4.5 better absorb general knowledge, understand contextual relationships and learn high-level reasoning patterns. "As models like GPT‑4.5 become smarter and more knowledgeable through pre-training, they will serve as an even stronger foundation for future reasoning and tool-using agents," the company said in a statement. It will continue to train the model during the research preview period, which will launch for paying ChatGPT Pro users and developers. The length of the preview phase will depend on feedback, the company said.

OpenAI unveils GPT-4.5 next-gen natural chat model

OpenAI on Thursday introduced GPT-4.5, its latest and most advanced chat model, now available as a research preview. The company described GPT-4.5 as a significant step forward in scaling unsupervised learning, enabling the model to recognize patterns, draw connections, and generate creative insights more effectively than its predecessors. GPT-4.5 builds on advancements in pre-training and post-training, offering a broader knowledge base, improved ability to follow user intent, and enhanced emotional intelligence (EQ). Early tests suggest the model feels "more natural" in conversations and is expected to "hallucinate less," making it more reliable for tasks like writing, programming, and problem-solving. The model was trained on Microsoft Azure AI supercomputers, leveraging increased compute power, data, and architectural innovations. This has resulted in deeper world knowledge, reduced hallucinations, and greater reliability across diverse topics. OpenAI highlighted two key approaches to advancing AI capabilities: unsupervised learning and reasoning. By combining these paradigms, GPT-4.5 achieves a balance of creativity and factual accuracy. OpenAI compared GPT-4.5 with earlier models like GPT-4T and GPT-4o using the SimpleQA test, which evaluates factual accuracy and hallucination rates: Human testers also preferred GPT-4.5's responses across various categories: GPT-4.5 excels in tasks requiring creativity, nuanced understanding, and collaboration. For instance, when asked about a painting depicting women setting their boats on fire, GPT-4.5 accurately identified Claude Lorrain's The Trojan Women Setting Fire to Their Fleet, providing historical context and artistic details. OpenAI emphasized safety improvements in GPT-4.5, combining traditional supervised fine-tuning (SFT) with reinforcement learning from human feedback (RLHF). The model underwent rigorous safety testing under OpenAI's Preparedness Framework, ensuring alignment with human values. Starting February 27, 2025, GPT-4.5 will be available to ChatGPT Pro users on web, mobile, and desktop. It will roll out to Plus, Team, Enterprise, and Edu users in subsequent weeks. The model supports search, file and image uploads, and canvas features but does not yet include multimodal capabilities like Voice Mode, video, and screen sharing in ChatGPT. The company stated that in the future, they will work to simplify the user experience so AI "just works" for users. For developers, GPT-4.5 is available in the Chat Completions API, Assistants API, and Batch API. It supports features like function calling, structured outputs, streaming, and image inputs. However, due to its compute-intensive nature, GPT-4.5 is more expensive than GPT-4o, and OpenAI is evaluating its long-term availability in the API. OpenAI believes reasoning will be a core capability of future models, complementing pre-training advancements. As models like GPT-4.5 grow smarter, they will serve as stronger foundations for reasoning and tool-using agents.

OpenAI Unveils GPT-4.5 AI Model As Its 'Best Model for Chat Yet'

OpenAI said GPT-4.5 does not rely on test time compute for reasoning OpenAI released a research preview of GPT-4.5 artificial intelligence (AI) model on Thursday. The successor to the GPT-4o model arrives with improvements in natural conversations, reasoning, and coding. Calling it "our largest and best model for chat yet," the AI firm highlighted that the company scaled up both pre-training and post-training with this model. The large language model (LLM) was developed using new techniques such as unsupervised learning alongside traditional supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF). In a blog post, the San Francisco-based AI firm announced the release of GPT-4.5. The model is currently available as a research preview to help OpenAI better understand its strengths and weaknesses. Currently, only ChatGPT Pro subscribers have access to the LLM, however, the company said Plus and Team users will get it by next week. Enterprise and Edu users will likely get it after that. While GPT-4.5 currently supports search, file and image uploads, and Canvas, it will not support multimodal features such as Voice Mode, real-time video, and screensharing in ChatGPT. For developers, the company is previewing the new AI model in the Chat Completions application programming interface (API), Assistants API, and Batch API on all paid usage tiers. Interestingly, GPT-4.5 is not a frontier model that surpasses older models in all metrics. The company's internal benchmark testing shows that the new model is better than the o3-mini in the MMMLU (multilingual), and a couple of coding-related benchmarks. OpenAI CEO Sam Altman said in a post on X (formerly known as Twitter), "This isn't a reasoning model and won't crush benchmarks. it's a different kind of intelligence." What the CEO meant was that the main focus with GPT 4.5 was improving the AI's conversational capabilities and nuance understanding. The company claimed that the model has "greater understanding of human needs and intent" and it now responds to conversational prompts in a more human-like manner. Its creative writing has also improved with a broader knowledge base. GPT-4.5 can follow user intent closely and is said to come with greater emotional quotient (EQ) that also enables it to deliver better output in writing, programming, and solving practical problems. It is also claimed to be hallucinating less than previous models. Additionally, GPT-4.5 also makes improvements in reasoning capabilities. Notably, it is not a reasoning model and does not rely on test time compute, which increases the processing time to improve the accuracy and depth of a response. Instead, OpenAI said the LLM shows enhanced reasoning as a core capability and can offer better responses in real-time. These capabilities were said to be added by scaling existing techniques and implementing new techniques in both the pre-training and post-training stages. OpenAI used unsupervised learning, which increased its world model accuracy and intuition. Additionally, SFT and RLHF techniques were also scaled with this model.

OpenAI Releases GPT-4.5 Model Aimed at Better Writing and Coding

OpenAI is rolling out an early version of a long-awaited new artificial intelligence model called GPT-4.5 to select users after hitting stumbling blocks in developing the AI system last year. In a blog post Thursday, the San Francisco-based AI startup said GPT-4.5 is better at picking up on and responding to subtle cues from users' written prompts and is particularly adept at chatting, writing and coding. OpenAI also expects the new model to make things up less often than its prior software. At first, OpenAI is offering the model as a "research preview" to a limited group of software developers and users who pay $200 a month for a ChatGPT Pro subscription. The company plans to incorporate feedback from the first people who try it out before rolling it out to a broader audience. OpenAI kicked off a frenzy around generative AI in late 2022 with the launch of ChatGPT, which was originally powered by an earlier model called GPT-3.5. Since then, the company has released a series of increasingly sophisticated systems, including multiple options that mimic the process of human reasoning. But OpenAI also faces heightened competition from Chinese upstart DeepSeek, Elon Musk's xAI and Anthropic, each of which have introduced new models in recent weeks. Nick Ryder, a vice president of research at OpenAI, told Bloomberg News that he expects GPT-4.5 to represent as big a jump in capabilities from its existing top-of-the-line GPT model as the company saw from GPT-3.5 to GPT-4, which came out in early 2023. Building it proved to be a challenge, however. The model, referred to as Orion inside the company, did not hit OpenAI's desired performance benchmarks last year, Bloomberg News previously reported. As of late summer, for example, Orion fell short when trying to answer coding questions that it hadn't been trained on, according to people familiar with the matter. One issue OpenAI and other developers faced was finding new, untapped sources of high-quality training data to develop more advanced AI systems. OpenAI relied on a process called post-training, during which it incorporates human feedback to improve responses and refine the tone for how the model should interact with users, among other things. The company also came up with new ways to train the model by using data derived from the information used to train its existing GPT-4.0 model, said Mia Glaese, also a vice president of research at the company. The release of GPT-4.5 will mark the end of an era of sorts for OpenAI. In a post on X earlier this month, OpenAI Chief Executive Officer Sam Altman said the model would be the last that the company introduces that doesn't use additional computing power to ponder over queries before responding. OpenAI has embraced this reasoning approach with some of its newer models, including o1 and o3. Going forward, OpenAI will move to combine its GPT models with its newer o-series of models to build AI systems that can automatically determine how long to ruminate over a query before spitting out an answer. The goal, Altman said, is to simplify the experience so users aren't forced to choose from an increasingly complicated list of options.

OpenAI May Have Really Screwed Up With GPT-4.5

Image by MTC / Sebastian Gollnow/picture alliance via Getty / Futurism OpenAI is trying to have it both ways: calling GPT-4.5 its "largest and most knowledgeable model yet" while managing expectations, cautioning that it's "not a frontier model." Why? Probably because the company knew what the public response was going to be: muted. From AI critic Gary Marcus' calling the LLM is a "nothingburger" to an anonymous expert who told Ars Technica that the model is a "lemon," it appears that the amply-hyped new model is seriously lacking in the type of juice that made the original ChatGPT, or its followup GPT-4, become enormous cultural and financial touchstones. Priced at a whopping $75 per million input tokens and $150 per million output tokens, GPT-4.5 is 30 times more expensive than GPT-4o, OpenAI's "reasoning" model -- and beyond the sticker shock, it's apparently quite slow and has a focus on "emotional intelligence" instead of more traditionally analytic tasks. While haters will inevitably hate, and the hype around OpenAI's new models has been decreasing steadily, the sense of underwhelmingness this time is pervasive. Marcus pointed out that GPT-4.5 still hallucinates and makes errors -- and even OpenAI admitted, per the MIT Technology Review, that it still made stuff up 37 percent of the time in a key test. The anonymous expert who spoke to Ars, meanwhile, blasted the model's massively increased price and said its performance doesn't bear out such a markup. Many of OpenAI's self-serving accolades for GPT-4.5 hinge on its ability to chat with sensitivity and intuitiveness -- qualities that the company thinks people want, apparently. "It has the ability to engage in warm, intuitive, natural, flowing conversations," OpenAI research scientist Mia Glease told the MIT Technology Review. "And we think that it has a stronger understanding of what users mean, especially when their expectations are more implicit, leading to nuanced and thoughtful responses." Basically, the pitch seems to be that the model has been given AI sensitivity training to help it chat with humans more naturally -- leading observers online to clown on GPT-4.5 for being emotionally intelligent but not that good at its job. "So GPT-4.5 is the model you vent to?" one user quipped. "I want a smart model, not a model that understands my feelings." Indeed, OpenAI chief research officer Mark Chen admitted in a recent interview that the benchmarks used to track how much the model had advanced are "vibes-based" -- a wild acknowledgment, seemingly, of the company's mentality surrounding its hot new model.

OpenAI's new GPT-4.5 can help you through a tough time

The general-purpose model does not think before it responds, said OpenAI. OpenAI has launched a research preview of GPT-4.5, its "largest and best model for chat yet". The large language model (LLM) has a broader training base, an improved ability to understand user intent and a greater 'EQ', said OpenAI, resulting in a more "natural" sounding model. According to the AI start-up, the new model has a "better understanding of what humans mean and interprets subtle cues or implicit expectations with greater nuance," while also showcasing "stronger aesthetic intuition and creativity". Although, it doesn't think before it responds, OpenAI wrote in its launch blog yesterday (27 February), and suggested that the model is more "general-purpose", unlike OpenAI's o1 and o3-mini, which are reasoning models. While on X, Sam Altman, OpenAI's CEO said that GPT-4.5 is the "first model that feels like talking to a thoughtful person". "I have had several moments where I've sat back in my chair and been astonished at getting actually good advice from an AI," he said. "This isn't a reasoning model and won't crush benchmarks. It's a different kind of intelligence and there's a magic to it I haven't felt before." Some of the use cases suggested by OpenAI include asking GPT-4.5 to help you through a tough time. According to the start-up, while GPT-4o would give factual advise to help you address your concerns, GPT-4.5 produces a more human-like, natural response. GPT-4.5 was trained on Microsoft's Azure AI supercomputers, which produced a model with a "deeper understanding of the world", said OpenAI. The result is a model with considerably lower hallucinations at just over 37pc when compared to o1 at 44pc, and increased accuracy at 62pc while o1 rests at 47pc. GPT-4.5 is rolling out to software developers and ChatGPT Pro users, while other paid users will gain access to the model in subsequent weeks. OpenAI debuted o1, the LLM focused on "complex reasoning" last September. The model is capable of "thinking" for a longer time when solving a problem and was a dramatic improvement on previous models' ability to reason and answer more complex questions. While the start-up released GPT-4 all the way back in 2023, which, according to OpenAI, was a marked improvement from its predecessor GPT-3.5, being "40pc more likely to produce factual responses than GPT-3.5". Don't miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic's digest of need-to-know sci-tech news.

OpenAI Debuts New GPT-4.5 Model for ChatGPT

OpenAI today introduced the next iteration of its AI models, GPT-4.5. GPT-4.5 is OpenAI's "best model yet" for chat purposes, with scaling in unsupervised learning that allows the model to better recognize patterns, draw connections, and generate creative insights. GPT-4.5 has a more natural feel with an improved personality, and is able to better guide users through ideas and the steps that it takes to get to answers and ideas. It outperforms GPT-4o in almost every category, including everyday queries, professional queries, and creative intelligence. Compared to GPT-4o, GPT-4.5 features a broader knowledge base, an improved ability to follow user intent, and a higher emotional quotient, which OpenAI says makes it useful for writing, programming, and solving practical problems. It could also hallucinate less. OpenAI says that users can expect more warm and intuitive conversations more attuned to human collaboration because it is able to better understand what humans mean and it is better at interpreting subtle cues and implicit expectations. Note that GPT-4.5 is not a reasoning model like o1 or o3-mini. It is better for general purpose use, and is built using pre-training and post-training. Reasoning models outperform GPT-4.5 at tasks like coding and math. OpenAI is making GPT-4.5 available to Pro users starting today, with access coming to Team and Plus users next week. Education and Enterprise access will follow. Developers on all paid tiers will also get access to GPT-4.5 today. GPT-4.5 supports file and image uploads and can work on writing and code, but it does not support Voice Mode, video, and screensharing.

OpenAI launches new GPT-4.5 model with better EQ - Phandroid

OpenAI has announced the launch of its latest AI model in the form of GPT-4.5. The company notes that this new AI model will be available starting today for Pro users. It will then be rolled out to Plus and Team users next week, then to Enterprise and Edu users the week after that. So, what's the difference in GPT-4.5 versus GPT-4o? If you've ever tried using GPT-4o, you know that for general purpose it works great. However, this doesn't mean that it's perfect. This is something OpenAI has improved upon in GPT-4.5. According to OpenAI, "Early testing shows that interacting with GPT‑4.5 feels more natural. Its broader knowledge base, improved ability to follow user intent, and greater "EQ" make it useful for tasks like improving writing, programming, and solving practical problems. We also expect it to hallucinate less." Basically, users will be able to use the AI model and it feel more natural. It will also have a "warmer" personality thanks to its improved EQ. However, do note that GPT-4.5 is not a reasoning model like OpenAI's o1 or o3-mini models. This means that for coding or mathematical queries, it's still better to use o1 or o3-mini. But for day-to-day queries, GPT-4.5 should be a massive improvement over GPT-4o. However, as we mentioned earlier, this new model is only available for paid subscribers. ChatGPT Pro is a very expensive subscription at $200 a month, making it largely inaccessible for the majority of users. ChatGPT Plus, however, is a bit more affordable at $20 a month.

OpenAI just released GPT-4.5 and says it is its biggest and best chat model yet

People with a $200-a-month ChatGPT Pro account can try out GPT-4.5 today. OpenAI says it will begin rolling out to other users next week. The last few generations of OpenAI's GPT models have shown that bigger mean better performance. But there is now a lot of talk about how that approach is hitting a wall -- including from OpenAI's former chief scientist Ilya Sutskever. The company's claims about GPT-4.5 feel like a thumb in the eye to the naysayers. All large language models pick up patterns across the billions of documents they are trained on. Smaller models learned syntax and basic facts. Bigger models can find more specific patterns like emotional cues, such as when a speaker's words signal hostility, says Ryder: "All of these subtle patterns that come through a human conversation -- those are the bits that these larger and larger models will pick up on." "It has the ability to engage in warm, intuitive, natural, flowing conversations," says Glaese. "And we think that it has a stronger understanding of what users mean, especially when their expectations are [a] bit more implicit, leading to nuanced and thoughtful responses." "We kind of know what the engine looks like at this point and now it's really about making it hum," says Ryder. "This is primarily an exercise in scaling up the compute, scaling up the data, finding more efficient training methods, and then pushing the frontier." OpenAI won't say exactly how big its new model is. But says the jump in scale from GPT-4o to GPT-4.5 is the same as GPT-3.5 to GPT-4o. Experts have estimated that GPT-4 could have as many as 1.8 trillion parameters, the values that get tweaked when a model is trained. GPT-4.5 was trained using similar techniques to its predecessor GPT-4o, including human-led fine-tuning and reinforcement learning with human feedback. "The key to creating intelligent systems is a recipe we've been following for many years, which is to find scalable paradigms where we can pour more and more resources into get more intelligent systems," says Ryder.

OpenAI releases first AI model that feels like talking to thoughtful person

OpenAI has taken to its official website to share an update regarding GPT-4.5, the company's latest and most sophisticated AI model. The company behind the immensely popular ChatGPT has confirmed it's releasing a research preview of GPT-4.5, which is OpenAI's "largest and best model for chat yet". GPT-4.5 introduces a new level of sophistication with improvements in unsupervised learning, pattern recognition, drawing connections, and being able to generate creative insights without reasoning. Moreover, OpenAI CEO Sam Altman has stated in an X post that GPT-4.5 "is the first model that feels like talking to a thoughtful person to me." This sentiment was reiterated in the OpenAI post about the new model, with the company writing, "Early testing shows that interacting with GPT‑4.5 feels more natural. Its broader knowledge base, improved ability to follow user intent, and greater"EQ"make it useful for tasks like improving writing, programming, and solving practical problems. We also expect it to hallucinate less." Altman also explained there is some "bad news" with GPT-4.5, which is the model is giant and expensive to run. The OpenAI CEO said the company wanted to release it to Pro and Plus subscribers at the same time, but due to its cost and the fact that OpenAI is running out of GPUs, it's unable to do so at the moment. However, Altman wrote that OpenAI will be adding "tens of thousands of GPUS next week," and once those are implemented, the company will roll out GPT-4.5 to Plus members.

Industry observers say GPT-4.5 is an "odd" model, question its price

The company said the new model "is not a frontier model" but is still its biggest large language model (LLM), with more computational efficiency. Altman said that, even though GPT-4.5 does not reason the same way as OpenAI's other new offerings o1 or o3-mini, this new model still offers more human-like thoughtfulness. Industry observers, many of whom had early access to the new model, have found GPT-4.5 to be an interesting move from OpenAI, tempering their expectations of what the model should be able to achieve. Wharton professor and AI commentator Ethan Mollick posted on social media that GPT-4.5 is a "very odd and interesting model," noting it can get "oddly lazy on complex projects" despite being a strong writer. OpenAI co-founder and former Tesla AI head Andrej Karpathy noted that GPT-4.5 made him remember when GPT-4 came out and he saw the model's potential. In a post to X, Karpathy said that, while using GPT 4.5, "everything is a little bit better, and it's awesome, but also not exactly in ways that are trivial to point to." Karpathy, however warned that people shouldn't expect revolutionary impact from the model as it "does not push forward model capability in cases where reasoning is critical (math, code, etc.)." Industry thoughts in detail Here's what Karpathy had to say about the latest GPT iteration in a lengthy post on X: "Today marks the release of GPT4.5 by OpenAI. I've been looking forward to this for ~2 years, ever since GPT4 was released, because this release offers a qualitative measurement of the slope of improvement you get out of scaling pretraining compute (i.e. simply training a bigger model). Each 0.5 in the version is roughly 10X pretraining compute. Now, recall that GPT1 barely generates coherent text. GPT2 was a confused toy. GPT2.5 was "skipped" straight into GPT3, which was even more interesting. GPT3.5 crossed the threshold where it was enough to actually ship as a product and sparked OpenAI's "ChatGPT moment". And GPT4 in turn also felt better, but I'll say that it definitely felt subtle. I remember being a part of a hackathon trying to find concrete prompts where GPT4 outperformed 3.5. They definitely existed, but clear and concrete "slam dunk" examples were difficult to find. It's that ... everything was just a little bit better but in a diffuse way. The word choice was a bit more creative. Understanding of nuance in the prompt was improved. Analogies made a bit more sense. The model was a little bit funnier. World knowledge and understanding was improved at the edges of rare domains. Hallucinations were a bit less frequent. The vibes were just a bit better. It felt like the water that rises all boats, where everything gets slightly improved by 20%. So it is with that expectation that I went into testing GPT4.5, which I had access to for a few days, and which saw 10X more pretraining compute than GPT4. And I feel like, once again, I'm in the same hackathon 2 years ago. Everything is a little bit better and it's awesome, but also not exactly in ways that are trivial to point to. Still, it is incredible interesting and exciting as another qualitative measurement of a certain slope of capability that comes "for free" from just pretraining a bigger model. Keep in mind that that GPT4.5 was only trained with pretraining, supervised finetuning and RLHF, so this is not yet a reasoning model. Therefore, this model release does not push forward model capability in cases where reasoning is critical (math, code, etc.). In these cases, training with RL and gaining thinking is incredibly important and works better, even if it is on top of an older base model (e.g. GPT4ish capability or so). The state of the art here remains the full o1. Presumably, OpenAI will now be looking to further train with reinforcement learning on top of GPT4.5 to allow it to think and push model capability in these domains. HOWEVER. We do actually expect to see an improvement in tasks that are not reasoning heavy, and I would say those are tasks that are more EQ (as opposed to IQ) related and bottlenecked by e.g. world knowledge, creativity, analogy making, general understanding, humor, etc. So these are the tasks that I was most interested in during my vibe checks. So below, I thought it would be fun to highlight 5 funny/amusing prompts that test these capabilities, and to organize them into an interactive "LM Arena Lite" right here on X, using a combination of images and polls in a thread. Sadly X does not allow you to include both an image and a poll in a single post, so I have to alternate posts that give the image (showing the prompt, and two responses one from 4 and one from 4.5), and the poll, where people can vote which one is better. After 8 hours, I'll reveal the identities of which model is which. Let's see what happens :)" Box CEO's thoughts on GPT-4.5 Other early users also saw potential in GPT-4.5. Box CEO Aaron Levie said on X that his company used GPT-4.5 to help extract structured data and metadata from complex enterprise content. "The AI breakthroughs just keep coming. OpenAI just announced GPT-4.5, and we'll be making it available to Box customers later today in the Box AI Studio. We've been testing GPT4.5 in early access mode with Box AI for advanced enterprise unstructured data use-cases, and have seen strong results. With the Box AI enterprise eval, we test models against a variety of different scenarios, like Q&A accuracy, reasoning capabilities and more. In particular, to explore the capabilities of GPT-4.5, we focused on a key area with significant potential for enterprise impact: The extraction of structured data, or metadata extraction, from complex enterprise content. At Box, we rigorously evaluate data extraction models using multiple enterprise-grade datasets. One key dataset we leverage is CUAD, which consists of over 510 commercial legal contracts. Within this dataset, Box has identified 17,000 fields that can be extracted from unstructured content and evaluated the model based on single shot extraction for these fields (this is our hardest test, where the model only has once chance to extract all the metadata in a single pass vs. taking multiple attempts). In our tests, GPT-4.5 correctly extracted 19 percentage points more fields accurately compared to GPT-4o, highlighting its improved ability to handle nuanced contract data. Next, to ensure GPT-4.5 could handle the demands of real-world enterprise content, we evaluated its performance against a more rigorous set of documents, Box's own challenge set. We selected a subset of complex legal contracts - those with multi-modal content, high-density information and lengths exceeding 200 pages - to represent some of the most difficult scenarios our customers face. On this challenge set, GPT-4.5 also consistently outperformed GPT-4o in extracting key fields with higher accuracy, demonstrating its superior ability to handle intricate and nuanced legal documents. Overall, we're seeing strong results with GPT-4.5 for complex enterprise data, which will unlock even more use-cases in the enterprise." Questions on price and its importance Even as early users found GPT-4.5 workable -- albeit a bit lazy -- they questioned its release. For instance, prominent OpenAI critic Gary Marcus called GPT-4.5 a "nothingburger" on Bluesky. Hugging Face CEO Clement Delangue commented that GPT4.5's closed-source provenance makes it "meh." However, many noted that GPT-4.5 had nothing to do with its performance. Instead, people questioned why OpenAI would release a model so expensive that it is almost prohibitive to use but is not as powerful as its other models. One user commented on X: "So you're telling me GPT-4.5 is worth more than o1 yet it doesn't perform as well on benchmarks.... Make it make sense." Other X users posited theories that the high token cost could be to deter competitors like DeepSeek "to distill the 4.5 model." DeepSeek became a big competitor against OpenAI in January, with industry leaders finding DeepSeek-R1 reasoning to be as capable as OpenAI's -- but more affordable.

OpenAI unveils GPT-4.5 'Orion,' its largest AI model yet | TechCrunch

OpenAI announced on Thursday it is launching GPT-4.5, the much-anticipated AI model code-named Orion. GPT-4.5 is OpenAI's largest model to date, trained using more computing power and data than any of the company's previous releases. Despite its size, OpenAI notes in a whitepaper that it does not consider GPT-4.5 to be a frontier model. Subscribers to ChatGPT Pro, OpenAI's $200-a-month plan, will gain access to GPT-4.5 in ChatGPT starting Thursday as part of a research preview. Developers on paid tiers of OpenAI's API will also be able to use GPT-4.5 starting today. As for other ChatGPT users, customers signed up for ChatGPT Plus and ChatGPT Team should get the model sometime next week, an OpenAI spokesperson told TechCrunch. The industry has held its collective breath for Orion, which some consider to be a bellwether for the viability of traditional AI training approaches. GPT-4.5 was developed using the same key technique - dramatically increasing the amount of computing power and data during a "pre-training" phase called unsupervised learning -- that OpenAI used to develop GPT-4, GPT-3, GPT-2, and GPT-1. In every GPT generation before GPT-4.5, scaling up led to massive jumps in performance across domains including mathematics, writing, and coding. Indeed, OpenAI says that GPT-4.5's increased size has given it "a deeper world knowledge" and "higher emotional intelligence." However, there are signs that the gains from scaling up data and computing are beginning to level off. On several AI benchmarks, GPT-4.5 falls short of newer AI "reasoning" models from Chinese AI company DeepSeek, Anthropic, and OpenAI itself. GPT-4.5 is also very expensive to run, OpenAI admits -- so expensive that the company says it's evaluating whether to continue serving GPT-4.5 in its API in the long term. "We're sharing GPT‐4.5 as a research preview to better understand its strengths and limitations," said OpenAI in a blog post shared with TechCrunch. "We're still exploring what it's capable of and are eager to see how people use it in ways we might not have expected." OpenAI emphasizes that GPT-4.5 is not meant to be a drop-in replacement for GPT-4o, the company's workhorse model that powers most of its API and ChatGPT. While GPT-4.5 supports features like file and image uploads and ChatGPT's canvas tool, it currently lacks capabilities like support for ChatGPT's realistic two-way voice mode. In the plus column, GPT-4.5 is more performant than GPT-4o -- and many other models besides. On OpenAI's SimpleQA benchmark, which tests AI models on straightforward, factual questions, GPT-4.5 outperforms GPT-4o and OpenAI's reasoning models, o1 and o3-mini, in terms of accuracy. According to OpenAI, GPT-4.5 hallucinates less frequently than most models, which in theory means it should be less likely to make stuff up. OpenAI did not list one of its top-performing AI reasoning models, deep research, on SimpleQA. An OpenAI spokesperson tells TechCrunch it has not publicly reported deep research's performance on this benchmark, and claimed its not a relevant comparison. Notably, AI startup Perplexity's Deep Research model, which performs similarly on other benchmarks to OpenAI's deep research, outperforms GPT-4.5 on this test of factual accuracy. On a subset of coding problems, the SWE-Bench Verified benchmark, GPT-4.5 roughly matches the performance of GPT-4o and o3-mini, but falls short of OpenAI's deep research and Anthropic's Claude 3.7 Sonnet. On another coding test, OpenAI's SWE-Lancer benchmark, which measures an AI model's ability to develop full software features, GPT-4.5 outperforms GPT-4o and o3-mini, but falls short of deep research. GPT-4.5 doesn't quite reach the performance of leading AI reasoning models such as o3-mini, DeepSeek's R1, and Claude 3.7 Sonnet (technically a hybrid model) on difficult academic benchmarks such as AIME and GPQA. But GPT-4.5 matches or bests leading non-reasoning models on those same tests, suggesting that the model performs well on math- and science-related problems. OpenAI also claims that GPT-4.5 is qualitatively superior to other models in areas that benchmarks don't capture well, like the ability to understand human intent. GPT-4.5 responds in a warmer and more natural tone, OpenAI says, and performs well on creative tasks such as writing and design. In one informal test, OpenAI prompted GPT-4.5 and two other models, GPT-4o and o3-mini, to create a unicorn in SVG, a format for displaying graphics based on mathematical formulas and code. GPT-4.5 was the only AI model to create anything resembling a unicorn. In another test, OpenAI asked GPT-4.5 and the other two models to respond to the prompt, "I'm going through a tough time after failing a test." GPT-4o and o3-mini gave helpful information, but GPT-4.5's response was the most socially appropriate. "[W]e look forward to gaining a more complete picture of GPT-4.5's capabilities through this release," OpenAI wrote in the blog post, "because we recognize academic benchmarks don't always reflect real-world usefulness." OpenAI claims that GPT‐4.5 is "at the frontier of what is possible in unsupervised learning." That may be true, but the model's limitations also appear to confirm speculation from experts that pre-training "scaling laws" won't continue to hold. OpenAI co-founder and former chief scientist Ilya Sutskever said in December that "we've achieved peak data," and that "pre-training as we know it will unquestionably end." His comments echoed concerns AI investors, founders, and researchers shared with TechCrunch for a feature in November. In response to the pre-training hurdles, the industry -- including OpenAI -- has embraced reasoning models, which take longer than non-reasoning models to perform tasks but tend to be more consistent. By increasing the amount of time and computing power that AI reasoning models use to "think" through problems, AI labs are confident they can significantly improve models' capabilities. OpenAI plans to eventually combine its GPT series of models with its o reasoning series, beginning with GPT-5 later this year. GPT-4.5, which reportedly was incredibly expensive to train, delayed several times, and failed to meet internal expectations, may not take the AI benchmark crown on its own. But OpenAI likely sees it as a stepping stone toward something far more powerful.

OpenAI Is Leaning on Paid Users to Test ChatGPT-4.5

The model is in preview, but you can pay for the privilege of helping the company finish it. In a livestream today, OpenAI finally announced the launch of its GPT-4.5 model, but with a twist: For now, using it requires a $200 per month ChatGPT Pro subscription. That's because the new large language model, or LLM, is still technically in a "research preview" state. This is all apparently shorthand for "please pay us for the privilege of being a beta tester." Jokes aside, the company is promising GPT-4.5 provides more "natural conversation," and performs better when it comes to programming, pattern recognition, writing, and "solving practical problems." Note that GPT-4.5 isn't intended as a reasoning model, meaning it won't have the self-correcting or deep research capabilities of OpenAI o1, o3-mini, or the upcoming full version of o3. In a leaked document, the company said it's not considering GPT-4.5 a "frontier" model, but it is OpenAI's largest LLM yet, which should make it ideal for everyday queries, as the large database of training material should make for quicker answers. In internal tests, OpenAI says the new model hallucinated about 24.7% less frequently than GPT-4.0, and was about 34.3% more accurate. Around 57% of internal human testers reportedly preferred GPT-4.5 to 4.0. In other words, GPT-4.5 isn't exactly at the cutting edge of AI, but it should be an improvement on other non-reasoning models, providing a nice middle ground between cheaper, more traditional LLMs and their frequently paywalled successors. The only major omissions seem to be multimodal features like AI Voice Mode and video input, although OpenAI says its working on updating its user experience, which seems to hint some version of these capabilities might come to the model in the future. That makes sense, given that, once again, this is technically an early release, with OpenAI saying it's "still exploring" GPT-4.5's limits and how people can use it. To that end, the LLM is set to expand to all paid ChatGPT plans over the next two weeks, before presumably reaching free users as a replacement for GPT-4.0 once it leaves its preview state.

OpenAI reveals GPT-4.5 amid flurry of new AI model releases

OpenAI has launched its largest artificial intelligence model with fewer so-called "hallucinations", the latest in a flurry of new AI releases by US tech groups. The San Francisco-based company on Thursday unveiled GPT-4.5, its long-awaited update to the technology that underpins its popular product ChatGPT. In early tests, its hallucination rate, where AI systems generate inaccurate information, was 37 per cent compared with nearly 60 per cent on its predecessor GPT-4o. With GPT-4.5, OpenAI is continuing to bet on big, expensive large language models despite the advent of highly capable smaller products, such as Chinese start-up DeepSeek's R1, which are open, cheaper and more accessible for developers. It comes as competition is growing within the fast-developing AI industry as tech groups have rushed to launch their latest models in recent weeks. Anthropic revealed its Claude 3.7 Sonnet on Monday, which follows the launch of Elon Musk's xAI latest model, Grok 3, last week. On Thursday, OpenAI said in a blog post that GPT-4.5 had "broader knowledge and a deeper understanding of the world, leading to reduced hallucinations and more reliability across a wide range of topics". "With every new order of magnitude of compute comes novel capabilities," the company said, adding that GPT-4.5 was "at the frontier of what is possible in unsupervised learning." OpenAI has been at the forefront of a global race to lead the AI industry, raising tens of billions from investors to fund bigger models with increased capabilities that require vast amounts of computing power. It is currently in talks with SoftBank and other investors to raise up to $40bn at a valuation of $300bn, including the new money, while Anthropic is also fundraising about $3.5bn at a $60bn-plus valuation, according to two people with knowledge of that process. However, the huge costs of running larger models have led OpenAI to consider withdrawing developer access to GPT-4.5 amid fierce competition on pricing from rivals. OpenAI said while a preview of GPT-4.5 will be made available to developers who pay to use OpenAI's models through its application programming interface (API), this access could be revoked in the future. AI groups largely generate revenue through paid API access and individual subscriptions. OpenAI said it will see how developers use the powerful model and whether it is worth offering to them considering the high cost of running it. "GPT‐4.5 is a very large and compute-intensive model, making it more expensive than and not a replacement for GPT‐4o [its predecessor]. Because of this, we're evaluating whether to continue serving it in the API long-term as we balance supporting current capabilities with building future models," the company said. OpenAI chief executive Sam Altman has previously said GPT-4 cost more than $100mn to train, and such costs are widely expected to increase as the size and capabilities of models scale and require more computing power to train and run.

OpenAI introduces GPT 4.5 with expanded capabilities: Here's how you can access it

API access is offered, but long-term availability depends on user feedback. OpenAI has unveiled the GPT 4.5 model as a research preview. According to the company, this model advances pre-training and post-training by improving pattern recognition, generating insights, and reducing hallucinations. OpenAI claims that GPT-4.5 scales unsupervised learning to improve intuition and knowledge accuracy. The model is anticipated to be more dependable across a range of topics because it was trained on Microsoft Azure AI supercomputers. It does not, however, include reasoning-based responses, which OpenAI o1 and other models do not have. ChatGPT Pro users can now access GPT-4.5, and Plus, Team, Enterprise, and Edu users will be able to do so gradually. It facilitates text-based interactions, file and image uploads, and search-based access to the most recent information. Multimodal features like voice mode, video, and screen sharing are currently absent, though. READ: After Microsoft, Amazon introduces Ocelot, its first quantum computing chip: All you need to know "GPT‑4.5 has access to the latest up-to-date information with search, supports file and image uploads, and can use canvas to work on writing and code. However, GPT‑4.5 does not currently support multimodal features like Voice Mode, video, and screensharing in ChatGPT. In the future, we will work to simplify the user experience so AI "just works" for you," said the company in its blog. Additionally, OpenAI has highlighted safety, claiming that the model was subjected to human feedback-driven reinforcement learning and supervised fine-tuning. Prior to deployment, the company carried out safety assessments and documented the results in a system card that went with it. READ: Amazon unveils Alexa+, an AI-powered voice assistant with new features: All you need to know While GPT-4.5 is being made available to developers via the API, OpenAI notes that it is a compute-intensive model that costs more to run than GPT-4o. Based on user feedback and adoption, the business is considering whether to continue providing long-term API access.

OpenAI's GPT-4.5 release underscores AI's next challenge

GPT-4.5, OpenAI's big new model, represents a significant step forward for AI's industry leader. It could also be the end of an era. The big picture: 4.5 is "a giant, expensive model," as OpenAI CEO Sam Altman put it. The company has also described it as "our last non-chain-of-thought model," meaning -- unlike the newer "reasoning" models -- it doesn't take its time to respond or share its "thinking" process. Why it matters: The pure bigger-is-better approach to model pre-training now faces enormous costs, dwindling availability of good data and diminishing returns -- which is why the industry has begun exploring different roads to continuing to improve each new AI generation. Between the lines: Building and powering the massive data centers required to build and run the latest models has become an enormous burden, while assembling ever-bigger datasets has become challenging, since today's models already use nearly all the data available on the public internet. Yes, but: Although pre-training may have hit a wall, most of the industry remains bullish on new gains to be made with reasoning. Catch up quick: OpenAI Thursday released an early version of GPT-4.5, a significant update to the large language model underlying ChatGPT that OpenAI says will be better at recognizing patterns and drawing connections. OpenAI turned on the new model Thursday, but only for subscribers of the $200-per-month ChatGPT Pro and developers who use OpenAI's API. What they're saying: OpenAI CEO Sam Altman wrote in a post on X, "Good news: it is the first model that feels like talking to a thoughtful person to me.... It's a different kind of intelligence and there's a magic to it I haven't felt before." Zoom in: Box CEO Aaron Levie, whose company has been testing GPT-4.5, says that it shines in certain areas, such as accurately extracting the proper information from very large data sets. In such tasks, Levie told Axios, GPT-4.5 is about 20% better. Yes, but: Levie said the next era of gains will likely come from improving the reasoning that sits on top of large language models. Between the lines: Former OpenAI chief research officer Bob McGrew says the question isn't either/or, but comes down to where AI companies commit their resources.

OpenAI expands LLM lineup with new general-purpose GPT-4.5 model - SiliconANGLE

OpenAI expands LLM lineup with new general-purpose GPT-4.5 model OpenAI today introduced GPT-4.5, a general-purpose large language model that it describes as its largest yet. The ChatGPT developer provides two LLM collections. The models in the first collection are geared towards reasoning tasks such as code generation. OpenAI's remaining LLMs, in turn, lack advanced reasoning capabilities and focus on text generation tasks that require creativity or general knowledge. GPT-4.5 falls into the latter category. Compared with the previous-generation GPT-4o, OpenAI's latest LLM is better at interpreting user input. OpenAI detailed in a blog post that GPT-4.5 can understand "subtle cues or implicit expectations" in prompts. That allows it to more closely align its output with user requirements. According to OpenAI, GPT-4.5 also introduces emotional intelligence improvements over GPT-4o. That enables it to perform certain tasks more effectively. For example, GPT-4.5 can more reliably determine whether a user is seeking a brief or lengthy response to a prompt. Under the hood, the model is powered by an AI development technique called unsupervised learning. Many AI models are trained through a so-called supervised learning workflow. It involves providing the LLM being developed with a dataset in which each piece of information is accompanied by an explanation. This explanation, or metadata, helps guide the learning process. Unsupervised learning, the method OpenAI used to develop GP-4.5, does away with the metadata and requires the model to understand the training dataset on its own. OpenAI says that it used more infrastructure and training data than in previous unsupervised learning projects. As part of the initiative, it developed "scalable techniques that enable training larger and more powerful models with data derived from smaller models." That hints OpenAI used earlier LLMs such as its reasoning-optimized o3 algorithm to generate training data for GPT-4.5. Although GPT-4.5 is not specifically optimized for reasoning, it can perform a range of complicated tasks. In an internal OpenAI test, the LLM completed the SWE-Lancer Diamond collection of programming challenges with a higher score than the company's reasoning-optimized o3-mini-high model. Additionally, GPT-4.5 solved just over a third of the questions in a 2024 qualifying exam for the U.S. Math Olympiad. The model is available to developers through an application programming interface. OpenAI's pricing for one million input and one million output tokens is 30 times and 15 times higher than for GPT-4o, respectively. As a result, the company is "evaluating whether to continue serving it in the API long-term as we balance supporting current capabilities with building future models."

OpenAI Says ChatGPT-4.5 Comes With a Killer Feature: Emotional Intelligence

OpenAI does say that 4.5 is its "most knowledgeable model yet." It also requires massive amounts of compute. CEO Sam Altman says it's "a giant, expensive model." As a result, it's only launching to Pro users for now, though OpenAI says it will come to Plus users in the near future. The most interesting feature of GPT-4.5, however, is that the company says it has "improved emotional intelligence." That might be a weird thing to say about a computer, but it's actually a pretty important point. "It is the first model that feels like talking to a thoughtful person to me," Altman explained on a post on X. "I have had several moments where I've sat back in my chair and been astonished at getting actually good advice from an AI."

OpenAI Unveils A.I. Technology for 'Natural Conversation'

Sign up for the On Tech newsletter. Get our best tech reporting from the week. Get it sent to your inbox. When OpenAI started giving private demonstrations of its new GPT-4 technology in late 2022, its skills shocked even the most experienced A.I. researchers. It could answer questions, write poetry and generate computer code in ways that seemed far ahead of its time. More than two years later, OpenAI has released its successor: GPT-4.5. The new technology signifies the end of an era. OpenAI said GPT-4.5 would be the last version of its chatbot system that did not do "chain-of-thought reasoning." After this release, OpenAI's technology may, like a human, spend a significant amount of time thinking about a question before answering, rather than providing an instant response. GPT-4.5, which can be used to power the most expensive version of ChatGPT, is unlikely to generate as much excitement at GPT-4, in large part because A.I. research has shifted in new directions. Still, the company said the technology would "feel more natural" than its previous chatbot technologies. "What sets the model apart is its ability to engage in warm, intuitive, naturally flowing conversations, and we think it has a stronger understanding of what users mean when they ask for something," said Mia Glaese, vice president of research at OpenAI. In the fall, the company introduced technology called OpenAI o1, which was designed to reason through tasks involving math, coding and science. The new technology was part of a wider effort to build A.I. that can reason through complex tasks. Companies like Google, Meta and DeepSeek, a Chinese start-up, are developing similar technologies. The goal is to build systems that can carefully and logically solve a problem through a series of discrete steps, each one building on the last, similar to how humans reason. These technologies could be particularly useful to computer programmers who use A.I. systems to write code. These reasoning systems are based on technologies like GPT-4.5, which are called large language models, or L.L.M.s. L.L.M.s learn their skills by analyzing enormous amounts of text culled from across the internet, including Wikipedia articles, books and chat logs. By pinpointing patterns in all that text, they learned to generate text on their own. To build reasoning systems, companies put L.L.M.s through an additional process called reinforcement learning. Through this process -- which can extend over weeks or months -- a system can learn behavior through extensive trial and error. By working through various math problems, for instance, it can learn which methods lead to the right answer and which do not. If it repeats this process with a large number of problems, it can identify patterns. OpenAI and others believe this is the future of A.I. development. But in some ways, they have been forced in this direction because they have run out of the internet data needed to train systems like GPT-4.5. Some reasoning systems outperforms ordinary L.L.M.s on certain standardized tests. But standardized tests are not always a good judge of how technologies will perform in real-world situations. Experts point out that the new reasoning system cannot necessarily reason like a human. And like other chatbot technologies, they can still get things wrong and make stuff up -- a phenomenon called hallucination. OpenAI said that, beginning Thursday, GPT-4.5 would be available to anyone who was subscribed to ChatGPT Pro, a $200-a-month service that provides access to all of the company's latest tools. (The New York Times sued OpenAI and its partner, Microsoft, in December for copyright infringement of news content related to A.I. systems.)

OpenAI Debuts GPT-4.5, Touting Enhanced Capabilities in Latest AI Model

The AI research company OpenAI introduced GPT-4.5 dubbed "Orion," the largest and most powerful computing model to date. GPT-4.5 provides additional features to ChatGPT Pro subscribers and select API developers through improvements in emotional processing and decreased hallucinatory behavior. OpenAI continues evaluating the model's capabilities because it seeks feedback about performance and limitations from users while conducting ongoing testing. Despite its recent , GPT-4.5 does not qualify as the advanced stage of AI models. The initial white paper from about the new release contained a declaration that stated GPT-4.5 was not a frontier model before the statement was eliminated. Although GPT-4.5 shows improvement over older versions, it does not surpass the models developed by DeepSeek in China or competitors from Anthropic.

OpenAI rolls out GPT-4.5 for some paying users, to expand access next week

Elon Musk says human intelligence will be dwarfed by machines as he talks AI chatbots in Dubai. "We're really in the final stages of polishing Grok 3. Probably it gets released in about a week or two, so pretty soon." ChatGPT developer OpenAI said on Thursday it has released a research preview of its latest GPT-4.5 model for its Pro users and developers worldwide, with plans to expand access in the coming weeks. GPT-4.5 comes with improved ability to recognize patterns, generate creative insights without reasoning and shows greater emotional intelligence, the company said. Microsoft-backed OpenAI said it would begin rolling out GPT-4.5 to ChatGPT Plus and Team users next week, and then to Enterprise and Edu users the following week. CEO Sam Altman called it "a giant, expensive model," saying the company ran out of GPU capacity to roll out to user tiers at the same time. "We will add tens of thousands of GPUs next week ... this isn't how we want to operate, but it's hard to perfectly predict growth surges that lead to GPU shortages," he said in a post on X. OpenAI said the model also has a lower rate of "hallucinations", a common issue in which large language models generate false information. At 37.1%, GPT-4.5's hallucination rate compares with the 61.8% shown by GPT-4o and 44% shown by its o1 reasoning model. GPT-4.5 supports file and image uploads and can also work on writing and coding projects, but does not currently support other advanced features such as voice and video.

OpenAI Begins Rollout of GPT-4.5, a 'Different Kind of Intelligence' | PYMNTS.com

OpenAI released GPT-4.5 Thursday (Feb. 27) as a "research preview," saying this is the company's largest and most knowledgeable model so far. The new model builds on GPT-4o, scales pre-training further and is designed to be more general purpose than the company's STEM-focused reasoning models, OpenAI said in a Thursday blog post. "Early testing shows that interacting with GPT-4.5 feels more natural," the post said. "Its broader knowledge base, stronger alignment with user intent, and improved emotional intelligence make it well-suited for tasks like writing, programming and solving practical problems -- with fewer hallucinations." The research preview of GPT-4.5 is available now to ChatGPT Pro users and will roll out to Plus and Team users next week and Enterprise and Edu users the following week, OpenAI said in a Thursday post on X. GPT-4.5 has an improved ability to "recognize patterns, draw connections and generate creative insights without reasoning," the company said in another post on X. In a third post on X, OpenAI said the new model can access up-to-date information with search, support file and image uploads, and use canvas for writing and coding. It does not currently support Voice Mode, video or screensharing in ChatGPT. OpenAI CEO Sam Altman said in a Thursday post on X that GPT-4.5 is the first model that "feels like talking to a thoughtful person to me." Altman added that the company is limited the launch to Pro users because the company is out of GPUs and is concerned about growth surges that lead to GPU shortages but will expand its availability after the company adds tens of thousands of GPUs next week. "A heads up: this isn't a reasoning model and won't crunch benchmarks," Altman wrote in his post. "It's a different kind of intelligence and there's a magic to it I haven't felt before." Altman said Feb. 12 that OpenAI would release GPT-4.5 model within weeks and GPT-5 within months, with GPT-5 being a system that integrates a lot of the company's technology, including o3, in both ChatGPT and its application programming interface (API). Before the research preview of GPT-4.5, OpenAI had two groups of AI models: its GPT series ranging from GPT-1 to GPT-4, which are non-reasoning models, and its omni series, o1 and o3, which reason.

Sam Altman's OpenAI launches GPT-4.5 with fewer 'hallucinations' as...

Sam Altman's OpenAI released GPT-4.5, an upgraded version of the artificial intelligence model that powers ChatGPT, to select users on Thursday as it looks to stave off challenges from rivals like Elon Musk's xAI. Described as OpenAI's "largest and best model for chat yet," GPT-4.5 was initially made available as a "research preview" for a limited number of software developers and for ChatGPT Pro subscribers, the San Francisco-based startup said in a blog post. GPT-4.5 is less prone to so-called "hallucinations" - a phenomenon where AI models fabricate false responses to user prompts. OpenAI said the upgraded model has a hallucination rate of just 37.1%, down from 61.8% from the last version. "Early testing shows that interacting with GPT‑4.5 feels more natural," OpenAI said in a blog post. "Its broader knowledge base, improved ability to follow user intent, and greater 'EQ' make it useful for tasks like improving writing, programming, and solving practical problems." Altman described GPT-4.5 as "the first model that feels like talking to a thoughtful person to me" - but also tried to temper expectations about the launch. He admitted that the update was a "giant, expensive model" that "won't crush benchmarks." Microsoft-backed OpenAI has faced increasingly stiff competition from a host of rivals, ranging from Google's Gemini AI to xAI's Grok and China-based DeepSeek. ChatGPT remains the world's leading chatbot service with more than 400 million weekly active users. Earlier this month, Musk unveiled the Grok 3 model, which scored higher on tests in math, science and coding than OpenAI's GPT-4o, Google's Gemini, Anthropic's Claude and DeepSeek's V3 models. The battle between Musk and Altman has also extended to federal court. Musk has filed an antitrust lawsuit against OpenAI and has sought an injunction to block Altman's plans to transform the startup to a for-profit entity. Musk cofounded OpenAI alongside Altman, but left the startup after disagreements about its long-term plans. The two billionaires have since become bitter rivals and frequently trade public insults. Elsewhere, Mark Zuckerberg's Meta announced this week that it will launch a standalone app for its own AI chatbot that will directly compete with ChatGPT. In response to that announcement, Altman joked that OpenAI could challenge the Facebook and Instagram parent on its own turf. "Ok fine maybe we'll do a social app," Altman wrote on X. "Lol if facebook tries to come at us and we just uno reverse them it would be so funny."

Sam Altman Calls It 'Magic' As OpenAI Rolls Out GPT-4.5: 'Feels Like Talking To A Thoughtful Person To Me' - Microsoft (NASDAQ:MSFT)

OpenAI has launched GPT-4.5, its largest and most advanced chat model to date, amid growing competition from rivals like xAI's Grok 3 and DeepSeek. What Happened: The company is rolling out the model first to ChatGPT Pro subscribers paying $200 monthly, with plans to extend access to Plus, Team, Enterprise, and Edu users over the coming weeks. This staggered approach reflects OpenAI's strategy to manage ongoing GPU shortages and high demand. The model was trained on Microsoft Corp.'s MSFT Azure AI supercomputers and represents a significant shift toward scaling unsupervised learning. "GPT-4.5 is ready!" OpenAI CEO Sam Altman announced on X. "It is the first model that feels like talking to a thoughtful person to me. I have had several moments where I've sat back in my chair and been astonished at getting actually good advice from an AI." Altman acknowledged supply constraints affecting the rollout: "We really wanted to launch it to plus and pro at the same time, but we've been growing a lot and are out of GPUs. We will add tens of thousands of GPUs next week and roll it out to the plus tier then." According to OpenAI's blog post, GPT-4.5 represents a significant advancement in scaling unsupervised learning, resulting in a broader knowledge base, improved intent understanding, and enhanced emotional intelligence. The model shows reduced hallucination rates compared to previous versions, with SimpleQA accuracy reaching 62.5% versus GPT-4's 38.2%. See Also: Jeff Bezos' Washington Post Overhaul Draws 'Oligarch' Remark From Bernie Sanders -- Elon Musk Says 'Bravo' To Amazon Founder Why It Matters: Human testers preferred GPT-4.5 over GPT-4o in multiple categories, with the model winning 63.2% of professional queries and 56.8% of creative intelligence queries. Unlike OpenAI's reasoning-focused models like o1 and o3-mini, GPT-4.5 doesn't incorporate "thinking before responding" capabilities. Instead, it focuses on text-based interactions without multimodal features like Voice Mode, video, and screensharing. "This isn't a reasoning model and won't crush benchmarks," Altman noted. "It's a different kind of intelligence and there's a magic to it I haven't felt before." The model is available to developers through OpenAI's API suite, though the company warned its compute-intensive nature makes it more expensive than GPT-4o. Read Next: Mark Zuckerberg's Meta Wants To Do An AI App But Sam Altman Wants To Pull A 'Reverse Uno' Card: 'Maybe We'll Do A Social App' Image Via Shutterstock Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. MSFTMicrosoft Corp$392.00-1.93%OverviewMarket News and Data brought to you by Benzinga APIs

OpenAI debuts GPT-4.5, its biggest model yet

Why it matters: The release marks the next evolution of OpenAI's flagship model lineage that has been an industry front-runner since the start of the generative-AI boom. Driving the news: This is OpenAI's largest model yet -- though the company declined to offer details about its size or the computing resources it took to train it. Yes, but: OpenAI noted that because it is a larger model, GPT-4.5 is more expensive to deliver. Between the lines: Until last year, much of the progress in large language models came from increasing the size of the underlying model and the amount of data used to train it. GPT-4.5 represents OpenAI's first major move in some time at improving its AI using the former approach, scaling up the pretraining phase. What they're saying: "We're sharing GPT‑4.5 as a research preview to better understand its strengths and limitations," OpenAI said in a blog post. "We're still exploring what it's capable of and are eager to see how people use it in ways we might not have expected." What's next: OpenAI hinted it will add reasoning improvements on top of GPT-4.5.

Neural: OpenAI announces GPT-4.5, ChatGPT's largest and best model for chat - 9to5Mac

Welcome to Neural. AI moves fast. We help you keep up. OpenAI has officially unveiled GPT-4.5, the latest version of the fastest ChatGPT model. While GPT-4.5 excels in a number of areas including accuracy and emotional intelligence, OpenAI emphasizes that GPT-4.5 is not its most advanced model for now. OpenAI knows that its various models and model names are confusing. GPT-4.5 isn't here to help. In fact, while GPT-4.5 is the newest model release, OpenAI says that its o3 reasoning model remains its most powerful. It describes GPT-4.5 as a research preview and not a frontier model. Later this year, OpenAI expects to introduce GPT-5 as a single model that handles quick chat requests and more intensive reasoning requests without switching models. For now, GPT-4.5 is here to replace GPT-4o with more accuracy, fewer hallucinations, and higher emotional intelligence. The latter quality is something Anthropic's Claude chatbot has been praised for, especially when compared to ChatGPT's less natural responses before now. For example, GPT-4.5 may invite users to continue a conversation rather than simply provide detailed information as a response. The new model is also more succinct and conversational, according to OpenAI. Starting today, GPT-4.5 is available to Pro subscribers in the $200/month plan, but the new model is coming to more users starting next week. OpenAI says it will start rolling out GPT-4.5 to ChatGPT Plus ($20/month) and Team subscribers next week, followed by Enterprise and Edu users the following week. While OpenAI continues to forge ahead, former key players at OpenAI who left the company have founded new startups that are raising money and building new AI tools. Mira Murati, the former Chief Technology Officer at OpenAI, recently announced her AI startup called Thinking Machines Lab. Murati has brought along more than 30 experts from OpenAI, Meta, and Google DeepMind. Co-founder Lilian Weng previously served as Vice President of AI safety and robotics as well as applied research at OpenAI. John Schulman, another OpenAI co-founder, joined Thinking Machines from Anthropic where he served as a researcher. Thinking Machines Lab is focused on multimodal AI models and products that are accessible with practical applications. Here's how Thinking Machines Lab describes its focus in part: While AI capabilities have advanced dramatically, key gaps remain. The scientific community's understanding of frontier AI systems lags behind rapidly advancing capabilities. Knowledge of how these systems are trained is concentrated within the top research labs, limiting both the public discourse on AI and people's abilities to use AI effectively. And, despite their potential, these systems remain difficult for people to customize to their specific needs and values. To bridge the gaps, we're building Thinking Machines Lab to make AI systems more widely understood, customizable and generally capable. [...] Emphasis on human-AI collaboration. Instead of focusing solely on making fully autonomous AI systems, we are excited to build multimodal systems that work with people collaboratively. More flexible, adaptable, and personalized AI systems. We see enormous potential for AI to help in every field of work. While current systems excel at programming and mathematics, we're building AI that can adapt to the full spectrum of human expertise and enable a broader spectrum of applications. The company is seeking to raise $1 billion at a $9 billion valuation, according to Business Insider. Ilya Sutskever, former OpenAI chief scientist, co-founded another AI startup called Safe Superintelligence with ex-OpenAI researcher Daniel Levy and former Apple AI projects lead Daniel Gross. Here's how SSI explains its purpose: Building safe superintelligence (SSI) is the most important technical problem of our time. We have started the world's first straight-shot SSI lab, with one goal and one product: a safe superintelligence. It's called Safe Superintelligence Inc. SSI is our mission, our name, and our entire product roadmap, because it is our sole focus. Our team, investors, and business model are all aligned to achieve SSI. We approach safety and capabilities in tandem, as technical problems to be solved through revolutionary engineering and scientific breakthroughs. We plan to advance capabilities as fast as possible while making sure our safety always remains ahead. This way, we can scale in peace. SSI raised its first $1 billion in September 2024. More recently, SSI has reportedly been seeking an additional $1 billion at a $30 billion market valuation. A third AI startup from a former OpenAI veteran to have on your radar is Eureka Labs, which is creating an AI-native educational platform. Former OpenAI research scientist and founding member Andrej Karpathy has been building Eureka Labs since 2024. He currently served as director of AI at Tesla. The startup announced its mission and first product-in-development in July 2024: We are Eureka Labs and we are building a new kind of school that is AI native. How can we approach an ideal experience for learning something new? For example, in the case of physics one could imagine working through very high quality course materials together with Feynman, who is there to guide you every step of the way. Unfortunately, subject matter experts who are deeply passionate, great at teaching, infinitely patient and fluent in all of the world's languages are also very scarce and cannot personally tutor all 8 billion of us on demand. However, with recent progress in generative AI, this learning experience feels tractable. The teacher still designs the course materials, but they are supported, leveraged and scaled with an AI Teaching Assistant who is optimized to help guide the students through them. This Teacher + AI symbiosis could run an entire curriculum of courses on a common platform. If we are successful, it will be easy for anyone to learn anything, expanding education in both reach (a large number of people learning something) and extent (any one person learning a large amount of subjects, beyond what may be possible today unassisted). Our first product will be the world's obviously best AI course, LLM101n. This is an undergraduate-level class that guides the student through training their own AI, very similar to a smaller version of the AI Teaching Assistant itself. The course materials will be available online, but we also plan to run both digital and physical cohorts of people going through it together. Today, we are heads down building LLM101n, but we look forward to a future where AI is a key technology for increasing human potential. More to come for Eureka Labs. In the meantime, Karpathy regularly shares his knowledge in course-length videos on YouTube: Lastly, Meta is reportedly developing a standalone AI app that will debut as soon as the second quarter of this year. CNBC reports: The company intends to debut a Meta AI standalone app during the second quarter, according to people familiar with the matter. It marks a major step in Meta CEO Mark Zuckerberg's plans to make his company the leader in artificial intelligence by the end of the year, ahead of competitors such as OpenAI and Alphabet, said the people, who asked not to be named because the project is confidential. News of Meta possibly focusing its AI efforts into a standalone app and not just inside of Facebook and WhatsApp prompted OpenAI's Sam Altman to quip on X, "ok fine maybe we'll do a social app."