San Francisco-based AI startup Anthropic has just released an upgraded version of its language model, Claude 2.1. This new model boasts an impressive 200,000-token context window, surpassing the recently released GPT-4 model from OpenAI, which only has a 120,000-token context window.

Anthropic’s partnership with Google has played a significant role in this upgrade. The expanded partnership provides Anthropic with advanced processing hardware, enabling Claude to handle much larger contexts.

With its enhanced capabilities, Claude 2.1 can now process lengthy documents such as full codebases or novels. This opens up new possibilities for applications ranging from contract analysis to literary study.

What sets Claude 2.1 apart is its ability to accurately grasp information from prompts that are over 50 percent longer than what GPT-4 can handle before performance degradation occurs. This represents a substantial improvement and showcases the model’s superior performance.

Another notable feature of Claude 2.1 is its 50 percent reduction in hallucination rates compared to its previous version, Claude 2.0. This increased accuracy brings Claude closer to competing with GPT-4 in responding precisely to complex factual queries.

Anthropic has also introduced new features, including an API tool for advanced workflow integration and “system prompts” that allow users to define Claude’s tone, goals, and rules at the outset for more personalized and contextually relevant interactions. For example, a financial analyst can direct Claude to adopt industry terminology when summarizing reports.

However, it’s important to note that the full 200,000-token capacity of Claude 2.1 is currently exclusive to paying Claude Pro subscribers. Free users will continue to have access to the 100,000-token capacity of Claude 2.0.

With its substantial context expansion and improved accuracy, Claude 2.1 is poised to compete head-to-head with leading models like GPT-4. Anthropic’s latest offering demonstrates its determination to provide cutting-edge AI capabilities.

(Image Credit: Anthropic)

See also: Paul O’Sullivan, Salesforce: Transforming work in the GenAI era

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with Digital Transformation Week.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The post Anthropic upsizes Claude 2.1 to 200K tokens, nearly doubling GPT-4 appeared first on AI News.

LEAVE A REPLY

Please enter your comment!
Please enter your name here