February 25, 2025

Claude 3.7 Sonnet support

Hassieb Pakzad

Langfuse says hello to Anthropic Claude 3.7 Sonnet with day 1 support for both the LLM playground, LLM-as-a-judge and cost tracking.

Anthropic has released their latest model, Claude 3.7 Sonnet as the first hybrid reasoning model on the market.

Core features include (from Anthropic release notes):

Hybrid Reasoning Capabilities: Claude 3.7 Sonnet is Anthropic’s first hybrid reasoning model, offering both near‑instant responses and an extended, step‑by‑step thinking mode with visible reasoning. API users can also set a token budget for how long the model thinks.
Enhanced Coding Skills: This release shows strong improvements in coding and front‑end web development, and it introduces Claude Code—a command line tool that lets developers delegate substantial engineering tasks directly from their terminal.
Unified, Real‑World Focus: Built with a philosophy that mirrors human dual thinking, Claude 3.7 Sonnet integrates ordinary LLM responses with deep reflection for improved performance on real‑world tasks, all while being available across all Claude plans at the same competitive pricing.

The cost of the new model remains unchanged from its predecessor, Claude 3.5 Sonnet.

Try it out

Was this page helpful?