AI models
Kimi K3.1 Pro vs Claude Opus 5.5
Same criteria, side by side. Where the gap is small, it may be measurement noise.
Also searched as
Kimi K3 vs Claude Opus
Direct answer: GPQA Diamond: Kimi K3.1 Pro 94.2%, Claude Opus 5.5 90.6% (Epoch AI). OTIS Mock AIME: Kimi K3.1 Pro 100%, Claude Opus 5.5 100% (Epoch AI). FrontierMath (tiers 1 to 3): Kimi K3.1 Pro 78.2%, Claude Opus 5.5 91.2% (Epoch AI). SimpleQA Verified: Kimi K3.1 Pro 62.2%, Claude Opus 5.5 72.2% (Epoch AI). GPQA Diamond (OpenRouter): Kimi K3.1 Pro 93.9%, Claude Opus 5.5 90.2% (OpenRouter Benchmarks). τ-bench Verified (airline): Kimi K3.1 Pro 78.7%, Claude Opus 5.5 75.3% (OpenRouter Benchmarks).
| Metric | Kimi K3.1 Pro | Claude Opus 5.5 | Source |
|---|---|---|---|
| Weights ?what it is | open | closed | no data |
| Input US$/1M ? | US$ 2.00 | US$ 4.00 | no data |
| Output US$/1M ? | US$ 10.00 | US$ 20.00 | no data |
| Context ? | 1,048,576 | 1,000,000 | no data |
| GPQA Diamond ? | 94.2% | 90.6% | Epoch AI |
| OTIS Mock AIME ? | 100% | 100% | Epoch AI |
| FrontierMath (tiers 1 to 3) ? | 78.2% | 91.2% | Epoch AI |
| SimpleQA Verified ? | 62.2% | 72.2% | Epoch AI |
| GPQA Diamond (OpenRouter) ? | 93.9% | 90.2% | OpenRouter Benchmarks |
| τ-bench Verified (airline) ? | 78.7% | 75.3% | OpenRouter Benchmarks |
Compare with
See lab: Moonshot AI · Anthropic
Sources
- Epoch AI, CC BY 4.0, latest version we checked.
- OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
- OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.
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.