OpenAI Unveils GPT-4o Mini: A Cost-Effective AI Model for Developers

OpenAI unveils cheaper version of top AI model, GPT-4o mini

OpenAI, the generative artificial intelligence startup backed by Microsoft (NASDAQ:MSFT), unveiled a smaller and cheaper version of its top end AI model on Thursday, known as GPT-4o mini. The new model requires less energy to run and costs 15 cents per million input tokens and 60 cents per million output tokens. That is more than 60% cheaper than GPT-3.5 Turbo, OpenAI said. OpenAI unveiled its flagship GPT-4o model in May. OpenAI said GPT-4o mini outperformed its rivals, getting a model score of 82% on Massive Multitask Language Understanding, compared to Google's (GOOG) (GOOGL) Gemini Flash, which scored a 77.9% and Anthropic's Claude Haiku, which scored a 73.8%. The company said the new model will be available later today for free users of ChatGPT, as well as those using ChatGPT Plus and Team. ChatGPT Enterprise users will get access next week. More on Microsoft The Rapidly Evolving World Of AI PCs Microsoft: Benefits From OpenAI Partnership Unveiling Microsoft's Long-Term Growth Plans Google, Microsoft allow Chinese companies access to Nvidia AI chips: report Microsoft Azure is 'stable,' Copilot should boost Office: BofA

ChatGPT Parent OpenAI Unveils GPT-4o Mini - A Cost-Effective AI Model For Developers - Microsoft (NASDAQ:MSFT)

OpenAI launches GPT-4o mini, a smaller, cheaper AI model.GPT-4o mini available now for free users and subscribers. Microsoft Corp MSFT backed OpenAI is introducing an affordable and compact version of its advanced AI model, aiming to attract a broader spectrum of developers and businesses. The new model, named GPT-4o mini, offers a cost-effective alternative to the highly capable GPT-4o. The San Francisco-based company said that GPT-4o mini will be available immediately for free users, as well as ChatGPT Plus and Team subscribers, with enterprise customers gaining access next week. This new model replaces GPT-3.5 Turbo in ChatGPT, Bloomberg reports. Originally unveiled in May, GPT-4o was designed to function as a superior digital assistant capable of processing audio and visual information in real time. Also Read: Whistleblowers Accuse OpenAI Of Hindering Staff From Reporting AI Risks GPT-4o mini retains much of this functionality, with more features planned for future updates. Developers have been testing GPT-4o mini over the past week. For example, email startup Superhuman used it to create automated email replies, and financial services startup Ramp utilized it to extract information from receipts. AI companies like Anthropic and Alphabet Inc's GOOGL Google often release smaller, less expensive versions of their top models to provide more options for developers, according to Bloomberg. Smaller models are suitable for automating basic tasks, while larger models handle more complex functions. This strategy allows developers to choose the most appropriate tool for their needs. Initially, GPT-4o mini can process and generate text and images, and it is the first model to use a new safety method called "instruction hierarchy," designed to prioritize certain instructions, such as those from a company, to prevent misuse. Read Next: ChatGPT-Parent OpenAI And Atomic Bomb-Fame Los Alamos Join Forces To Explore AI's Potential In Scientific Research Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Photo courtesy: Shutterstock Market News and Data brought to you by Benzinga APIs

OpenAI debuts mini version of its most powerful model yet

OpenAI CEO Sam Altman speaks during the Microsoft Build conference at Microsoft headquarters in Redmond, Washington, on May 21, 2024. OpenAI on Thursday launched a new AI model, "GPT-4o mini," the artificial intelligence startup's latest effort to expand use of its popular chatbot. The company called the new release "the most capable and cost-efficient small model available today," and it plans to integrate image, video and audio into it later. The mini AI model is an offshoot of GPT-4o, OpenAI's fastest and most powerful model yet, which it launched in May during a livestreamed event with executives. The o in GPT-4o stands for omni, and GPT-4o has improved audio, video and text capabilities, with the ability to handle 50 different languages with improved speed and quality, according to the company. OpenAI, backed by Microsoft, has been valued at more than $80 billion by investors. The company, founded in 2015, is under pressure to stay on top of the generative AI market while finding ways to make money as it spends massive sums on processors and infrastructure to build and train its models. The mini AI model announced Thursday is part of OpenAI's push to be at the forefront of "multimodality," or the ability to offer a wide range of types of AI-generated media, like text, images, audio and video, inside one tool: ChatGPT. Last year, OpenAI COO Brad Lightcap told CNBC, "The world is multimodal. If you think about the way we as humans process the world and engage with the world, we see things, we hear things, we say things - the world is much bigger than text. So to us, it always felt incomplete for text and code to be the single modalities, the single interfaces that we could have to how powerful these models are and what they can do." GPT-4o mini is available starting Thursday to free users of ChatGPT, along with ChatGPT Plus and Team subscribers, and it will be available to ChatGPT Enterprise users next week, the company said in a release.

How Microsoft-Backed OpenAI's 'Mini' AI Model Is Heating Up the AI Competition

As a major OpenAI investor, Microsoft could also potentially stand to benefit from OpenAI's gains from the small model. Microsoft-backed (MSFT) OpenAI unveiled GPT-4o mini, its latest artificial intelligence (AI) model that is smaller and lower cost than its predecessors, in a move that could help advance the AI pioneer's position in the AI race. GPT-4o mini offers some of the capabilities of OpenAI's larger models at a lower cost, and could appeal to those looking for AI systems for specific uses that don't require the same computing power as bigger models, which could attract new customers and expand the company's AI leadership. OpenAI claimed that GPT-4o mini outperformed GPT-3.5 Turbo, the company's previous task-specific model, as well as some competitors' small model offerings including Google parent Alphabet's (GOOGL) Gemini Flash and Amazon-backed (AMZN) Anthropic's Claude Haiku. While GPT-4o mini outperformed its small model peers in several benchmarks, it still lagged OpenAI's larger GPT-4o. The company did not compare GPT-4o mini's performance to that of Microsoft's small model, Phi-3 Mini. As a major OpenAI investor, Microsoft could potentially stand to benefit from OpenAI's gains from GPT-4o mini. GPT-4o mini's low cost and better performance compared to alternative small models could attract more customers to OpenAI. The small model could also be especially beneficial to OpenAI as an Apple (AAPL) partner as a new AI-supercharged iPhone model is expected in the coming months. The new model is available to ChatGPT Free, Plus, and Teams users as of Thursday, as well as developers. Enterprise users are set to have access starting next week.

Breaking: OpenAI Launches GPT-4o Mini as Latest AI Breakthrough

OpenAI addresses SEC whistleblower concerns, revising policies to bolster open reporting and discussion. OpenAI has announced the introduction of the GPT-4o Mini, a streamlined and more accessible version of its state-of-the-art AI model. This new release is an addition to OpenAI's contribution in the development of AI solutions with the aim of expanding their usage. On Thursday, OpenAI, an Artificial Intelligence research company that is backed by Microsoft, released the GPT-4o Mini to the public. Based on GPT-4o, which has "omni" capabilities across multiple formats, the new model is hailed as the most capable and affordable small model on the market today. The GPT-4o Mini has been rolled out to all ChatGPT users including those on the free, Plus, and Team plans, with the Enterprise clients being expected to get it next week.