Anthropic Set to Launch Advanced Hybrid AI Model with Variable Reasoning Capabilities

Anthropic Plans to Release New Hybrid AI Model in Coming Weeks: Reports

Anthropic, the company behind the Claude family of AI models, has developed a new hybrid model, The Information reported on Thursday. The hybrid AI model will use more computational resources for complex queries where reasoning is required and will quickly answer simpler tasks when needed. Anthropic plans to release it in the coming weeks, the report added. Moreover, this model is said to give customers control over how long it takes to reason to solve a query. As revealed in the report, developers will be able to adjust the tokens using a 'sliding scale'. In another report, The Information stated that Anthropic expects to grow its revenue to over $34 billion in 2027. In addition to new technology, the company is said to have burned $5.6 billion in cash last year and expects to stop the cash burn by 2027. On the other hand, even OpenAI has taken a similar approach to building a hybrid system, as revealed by CEO Sam Altman. GPT-4.5, internally referred to as Orion, will be the next release and the last non-chain-of-thought model. Moreover, OpenAI plans to unify the o-series and GPT-series models, enabling systems that can integrate all tools and determine optimal thinking time for tasks. GPT-5 will be introduced in ChatGPT, and the API will be introduced as a system incorporating various OpenAI technologies, including o3. The company will not release o3 as a standalone model. Moreover, Dario Amodei, CEO of Anthropic, revealed in November last year that the flagship Claude 3.5 Opus will eventually be released. According to reports, Anthropic has finished training Claude 3.5 Opus, which performed well and scaled appropriately. Recently, Anthropic unveiled a report to showcase the impact of AI on how people work. It revealed that roughly 36% of all occupations use AI for at least a quarter of their associated tasks. Moreover, 57% of use cases involved leveraging AI to augment human capabilities, while 43% suggested automation. However, only 4% of occupations showcased AI usage for at least 75% of their tasks. It was also reported that Google is investing over a billion dollars in Anthropic, in addition to the $2 billion round led by Lightspeed Venture Partners last month.

Report: Anthropic Set to Launch Hybrid AI Model With Variable Reasoning Levels | PYMNTS.com

Anthropic reportedly plans to launch within weeks a hybrid artificial intelligence (AI) model that includes reasoning capabilities. The model aims to give developers more control over the cost of using an AI model by enabling them to control the amount of computational resources it spends on a problem, Seeking Alpha reported Thursday (Feb. 13), citing a paywalled article by The Information. Anthropic's upcoming hybrid AI model uses more computational resources to answer hard questions and acts like a traditional large language model (LLM) to handle simple questions, according to the report. This cost-effective approach is meant to appeal to the enterprise market that the company aims to serve, the report said. Anthropic did not immediately reply to PYMNTS' request for comment. This report came about a week after OpenAI released what it called its "most cost-efficient" reasoning AI model, the o3-mini. The o3-mini is part of OpenAI's o1 series, which can reason through tasks but takes longer to respond than non-reasoning models. However, unlike the other models in that series, the o3-mini is "small" but "powerful and fast," OpenAI said. The model outperforms earlier models especially in science, coding and math, and comes in three reasoning levels: low, medium and high. It was reported Jan. 22 that Google was investing more than $1 billion in Anthropic, upping its stake after already giving the AI firm around $2 billion, and that Anthropic was also on the verge of landing another $2 billion from a group of venture capital investors. Together, these deals were expected to bring Anthropic's valuation to around $60 billion. On Feb. 6, Anthropic and ridesharing company Lyft said they are working together to create AI products to serve Lyft's riders and drivers. The first phase of the collaboration is already underway, with Lyft's customer care AI assistant being powered by Anthropic's AI model, Claude, via Amazon Bedrock. The collaboration will also include other AI applications designed to enhance the rideshare experience for both riders and drivers, Lyft's participation in early testing for new products and capabilities, and Anthropic's training of Lyft's engineering teams to accelerate their ability to build AI-powered features.

Anthropic's next major AI model could arrive within weeks | TechCrunch

AI startup Anthropic is gearing up to release its next major AI model, according to a report Thursday from The Information. The report describes Anthropic's upcoming model as a "hybrid" model, which can switch between "deep reasoning" and fast responses. The company will reportedly introduce a "sliding scale" alongside the model to allow developers to control costs, as the deep reasoning capabilities consume more computing. Anthropic's model, which could arrive within weeks, outperforms OpenAI's o3-mini-high "reasoning" model on some programming tasks, according to the report. The model is also said to excel at analyzing large codebases and other business-related benchmarks. Anthropic CEO Dario Amodei hinted at new models in a recent interview with TechCrunch's own Romain Dillet. "We're generally focused on trying to make our own take on reasoning models that are better differentiated," Amodei said. "We've been a little bit puzzled by the idea that there are normal models and there are reasoning models and that they're sort of different from each other."