Anthropic has launched its newest model, called the Claude 3.5 Sonnet, which promises to match or surpass OpenAI’s GPT-4 and Google’s Gemini on a wide range of tasks. This new template is already available to Claude users on the web and iOS, and Anthropic is making it accessible to developers as well. Claude 3.5 Sonnet will be positioned as the intermediate model in Anthropic’s range, which uses the name Haiku for its smallest model, Sonnet for the medium option and Opus for the flagship model.
The company claims that the Claude 3.5 Sonnet outperforms the 3 Opus, and benchmarks show that it does so by a fairly large margin. The new model is also apparently twice as fast as its predecessor. Anthropic claims that the Claude 3.5 Sonnet will be much more adept at writing and translating code, managing multistep workflows, interpreting graphs and tables, and transcribing text from images. This new and improved Claude is also apparently better at understanding humor and can write much more humanely.
Along with the new template, Anthropic is also introducing a new feature called Artifacts. With Artifacts, it will be possible to see and interact with the results of requests made to Claude: if you ask the model to design something, it can now show what he created and allow you to edit it directly in the app. If Claude writes an email, you can edit it in the Claude app instead of having to copy it into a text editor. Artifacts seems to be a sign of the long-term vision for Claude. Anthropic has always stated that it is primarily focused on businesses.
