Up to four models
Claude Opus 5 vs Kimi K3.1 Pro
On the index (GPT-4 = 100), Claude Opus 5 scores 185 and Kimi K3.1 Pro 179. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Opus 5
185169–201
WeightsClosed
ReleasedJul 2026
Price in / out$5.0 / $25
Context1M
Core tests taken6 of 9
Kimi K3.1 Pro
179163–195
WeightsOpen
ReleasedOct 2026
Price in / out$0.24 / $8.4
Context1.05M
Core tests taken4 of 9
Claude Opus 5 and Kimi K3.1 Pro are too close to call (6 points apart, within the uncertainty), ahead on 2 of 4 shared core tests, at 4.4× the price.
Test by testthird-party results
Claude Opus 5Kimi K3.1 Pro
Every result as a table
| Test | Claude Opus 5 | Kimi K3.1 Pro |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 93.9% | 94.2% |
| SimpleQA Verified | 59.9% | 62.2% |
| Maths | ||
| FrontierMath (tiers 1-3) | 85.6% | 78.3% |
| AIME-style maths (OTIS mock) | 98.9% | 100.0% |
| ProofBench v1.1 | 99.0% | 91.0% |
| Chess puzzles | 43.0% | 45.0% |
| FrontierMath (tier 4) | 73.2% | 65.8% |
| Coding | ||
| Terminal-Bench 4.0 | 53.5% | 63.6% |
| IOI | 84.3% | 81.9% |
| Vibe Code Bench v1.1 | 88.4% | 84.9% |
| Agents | ||
| APEX-Agents | 65.8% | 65.5% |
| BioMysteryBench | 79.3% | 75.6% |
| SRE Bench | 12.2% | 27.1% |
| Terminal-Bench-Science | 27.1% | 31.4% |
| DTBench | 97.9% | 93.3% |
| Furniture assembly | 60.8% | 55.8% |
| LMCA | 64.5% | 57.6% |
| Mystery game puzzles | 59.0% | 42.0% |
| Professional work | ||
| Legal Research Bench | 55.3% | 37.5% |
| Tax Agent Bench | 47.1% | 24.9% |
| Novel problems | ||
| MysteryMechanism | 37.4% | 26.1% |
Tests at least two of these models have taken. Bold rows are the core tests.