Up to four models
Claude Opus 4.6 vs Grok 4.7
On the index (GPT-4 = 100), Claude Opus 4.6 scores 165 and Grok 4.7 164. 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 4.6
165149–181
WeightsClosed
ReleasedFeb 2026
Price in / out$5.0 / $25
Context1M
Core tests taken7 of 9
Grok 4.7
164149–180
WeightsClosed
ReleasedSep 2026
Price in / out$2.0 / $6.0
Context500K
Core tests taken6 of 9
Claude Opus 4.6 and Grok 4.7 are too close to call (1 point apart, within the uncertainty), ahead on 2 of 4 shared core tests, at 3.3× the price.
Test by testthird-party results
Claude Opus 4.6Grok 4.7
Every result as a table
| Test | Claude Opus 4.6 | Grok 4.7 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 90.5% | 92.7% |
| SimpleQA Verified | 47.0% | 56.0% |
| Maths | ||
| FrontierMath (tiers 1-3) | 66.0% | 53.0% |
| AIME-style maths (OTIS mock) | 94.4% | 98.1% |
| Chess puzzles | 21.0% | 41.0% |
| FrontierMath (tier 4) | 26.8% | 17.1% |
| Coding | ||
| Vibe Code Bench v1.1 | 57.6% | 86.2% |
| MirrorCode | 33.3% | 54.4% |
| Agents | ||
| APEX-Agents | 46.3% | 70.0% |
| DTBench | 91.2% | 96.0% |
| Furniture assembly | 28.3% | 20.8% |
| LMCA | 55.8% | 49.4% |
| Mystery game puzzles | 25.0% | 29.0% |
| Professional work | ||
| LegalBench | 85.3% | 84.4% |
| MedCode | 49.1% | 49.5% |
| MedScribe | 86.7% | 89.4% |
| Novel problems | ||
| ARC-AGI-2 | 69.2% | 61.4% |
| ARC-AGI-1 | 94.0% | 90.2% |
Task length · METR47 sec
Tests at least two of these models have taken. Bold rows are the core tests.