The new Claude model of Anthropic may think both quickly and slowly
Another week and there is another new AI model ready for public use. This time is anthropic with the introduction of Claude 3.7 SonnetS The company describes its latest edition as the first market "a hybrid model for reasoning," The new version of Claude can answer a question almost instantly or take time to work through it step by step. As a user you can decide what approach the Claude is using, with a drop -down menu, which allows you to select "thinking mode" You want to take.
"We have developed CLUDE 3.7 Sonnet with different philosophy from other models for market reflections. Just as people use a brain for both quick answers and deep reflection, we believe that reasoning should be the integrated ability of border models, not a separate model entirely," He writes Anthrop. "This unified approach also creates a more problem experience for consumers."
Anthropic does not explicitly name Openai, but the company apparently takes a shot to its rival. Between GPT-4, O1, o1-mini And now O3-MiniOpenai offers many different models, but unless you follow the company closely, the number of systems offered can be huge; Actually Sam Altman recently admitted so muchS "We hate the pattern of the model as much as you and we want to go back to Magic Unified Intelligence," He Posted on x Earlier this month.
Anthropic says he also used a different approach to develop Claude’s reasoning opportunities. "We were optimized a little less about problems with competition in mathematics and computer science and instead directed the focus to real-world tasks that better reflect how the business actually uses LLMS," The company writes. Claude’s current users can be eagerly awaited by this time "Particularly strong improvements in the coding and development of the front end."
Claude 3.7 Sonnet is available for use today in all Claude plans, including the free level of Anthropic. Meanwhile, developers can gain access to the new model through the company’s API, Amazon Bedrock and Vertex AI on Google Cloud.
Speaking of developers, Anthropic also introduces Claude Code, new "agent" A tool that allows you to delegate encoding tasks to slaughter directly from the terminal interface. Available currently as a limited examination review, Anthropic says Claude Code can read code, edit files, write and launch tests, and even press engagements with Github.
This article originally appeared at Engadget at https://www.engadget.com/Ai/anthropics-new-claude-Model-can-tink-ast- and-and-203307140.html?src=ss
English