Anthropic releases a new powerful generative AI model called Claude 3.5 Sonnet. The developer claims that this is the best model among all of them. This was reported by TechCrunch.
Claude 3.5 Sonnet can generate and analyze text, and it can also analyze images. In several reading, coding, and math tests, Claude 3.5 Sonnet outperforms the previous Claude 3 Sonnet model, and outperforms Anthropic's previous flagship model, the Claude 3 Opus.
Benchmarks are not always the most useful indicator of AI performance, in part because many of them test edge cases that are unlikely to be used in widespread use, such as answering medical exam questions.
Still, in Anthropic's tests, Claude 3.5 Sonnet outperforms leading competitors' models, including OpenAI's recently launched GPT-4o.
Along with the new model, Anthropic is releasing so-called Artifacts, workspaces where users can edit and add content, such as code and documents generated by Anthropic models.
The context window of Claude 3.5 Sonnet is 200,000 tokens. Given that one token is not a whole word but its syllables, the context window of the new model will contain approximately 150,000 words.
According to Anthropic, Artifacts will soon receive new features such as sharing for large groups of users and database storage.
Anthropic claims that the new model better understands nuances and complex instructions, as well as things like humor. In addition, Claude 3.5 Sonnet is about twice as fast as Claude 3 Opus.
Anthropic's product director Michael Gerstenhaber says that these improvements are the result of architectural changes and new training data, including those generated by artificial intelligence.
“What matters to [businesses] is whether or not AI is helping them meet their business needs, not whether or not AI is competitive on a benchmark,” Gerstenhaber told TechCrunch. “And from that perspective, I believe Claude 3.5 Sonnet is going to be a step function ahead of anything else that we have available — and also ahead of anything else in the industry.”