Anthropic has released the Claude 3.7 Sonnet model, a significant step forward in the field of artificial intelligence, designed to deliver faster responses and enhanced reasoning capabilities. This move comes as the generative AI sector experiences a surge in competition, with US tech firms and companies from countries like China striving for advancements.
The new Sonnet model, backed by Amazon and Google, is available on all Claude plans, including Free, Pro, Team, and Enterprise. Anthropic highlights that the ‘extended thinking mode’ feature, available exclusively to paid subscribers, allows the model to ‘self-reflect before answering,’ thereby boosting its performance. This is particularly noticeable in tasks relating to mathematics, physics, coding, and instruction-following.
According to Anthropic, the San Francisco-based company, the hybrid reasoning model targets what they term as ‘real-world’ tasks. The goal is to create a more practical applicability for businesses using large language models, rather than prioritizing complex mathematical or computer science problems.
In addition to the Sonnet model, Anthropic is also offering a limited-release preview of Claude Code. This agentic coding tool assists developers with coding tasks, allowing them ‘to delegate substantial engineering work directly from their terminal.’ Agentic coding tools are essentially AI-powered applications designed to autonomously handle coding duties.
Despite these advancements, Anthropic maintains its competitive edge, announcing that its pricing structure will mirror that of previous models, thus providing cost-effective solutions. The Claude 3.7 Sonnet’s pricing is notably lower than that of its competitor, OpenAI’s o1 model, with costs of $3 per million input tokens and $15 per million output tokens, in contrast to OpenAI’s $15 and $60, respectively, solidifying its position in the market.
