Claude Sonnet 5 (Adaptive Reasoning)
FrontierClosed weightsTextAdaptive reasoning
Claude 5 family model by Anthropic.
About the Claude 5 family
Anthropic's 2026 generation: Opus 5 (most capable) and Sonnet 5 (balanced), both using adaptive reasoning — the model gauges how much thinking a problem needs — with effort levels from low to max.
All Claude 5 modelsHow this one is configured
This configuration uses adaptive reasoning (decides per-question how much to think).
Also runs at: low · medium · high · xhigh · max effort
Effort controls how long the model thinks before answering — higher effort usually means better answers on hard problems, at the cost of speed and money. Same model underneath, so they share this page.