Compare AI models head-to-head
Side-by-side comparisons of the most-asked-about AI models. Benchmarks (MMLU, GPQA, SWE-Bench, HumanEval), pricing, context windows, and capabilities — all in one view.
Featured matchups
claude-opus-4-7vsclaude-sonnet-4-6
Claude tier choice — Opus vs Sonnet.
Compare →deepseek-r1vsclaude-opus-4-6
Open-weights reasoning vs Claude.
Compare →kimi-k2vsclaude-sonnet-4-6
Kimi K2 vs Sonnet for agentic coding.
Compare →Don't see the comparison you want? Open any model detail page and click “Compare with…” — or construct a URL like
/models/compare/anthropic--claude-opus-4-7/openai--gpt-5.2.