GPQA Diamond
?what it ishighest score on this metric 96.4%
42º / 198 · Epoch AI, latest version we checked
Benchmarks, price, context and versions
Claude-Sonnet-5
Direct answer: Claude Sonnet 5 is a model from Anthropic. 90.5% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 2.00 per 1 million input tokens and US$ 10.00 per 1 million output tokens.
This block follows a rule of this section, using source data: if the model is in the top 25% of the metric, the item goes under "Good for"; if it is in the bottom half, it goes under "Not the best for". The middle band appears in neither. Price and context use OpenRouter; scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.
highest score on this metric 96.4%
42º / 198 · Epoch AI, latest version we checked
highest score on this metric 100%
46º / 186 · Epoch AI, latest version we checked
highest score on this metric 95.1%
37º / 86 · Epoch AI, latest version we checked
highest score on this metric 94%
60º / 87 · Epoch AI, latest version we checked
highest score on this metric 96.5%
59º / 135 · OpenRouter Benchmarks, latest version we checked
highest score on this metric 82%
47º / 131 · OpenRouter Benchmarks, latest version we checked
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.
Claude Sonnet 5 is a model from Anthropic, with closed weights, used for text, agents.
On OpenRouter, US$ 2.00 per 1 million input tokens and US$ 10.00 per 1 million output tokens.
1,000,000 tokens, per OpenRouter.
No. It is available through API only.
90.5% on GPQA Diamond, per Epoch AI.
No data: there is no SWE-bench Verified measurement for it in open sources.
It was made by Anthropic, a lab from the US.
provisional data Questions under review: Semrush had no API balance during collection. Answers use source data only.
Anthropic does not publish weights on Hugging Face: data only via OpenRouter and Epoch AI.
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.