Up to four models
Claude Opus 4.7 vs MiniMax-M2.7
On the index (GPT-4 = 100), Claude Opus 4.7 scores 167 and MiniMax-M2.7 132. 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 4.7
167151–183
WeightsClosed
ReleasedApr 2026
Price in / out$5.0 / $25
Context1M
Core tests taken6 of 9
132114–151
WeightsOpen
ReleasedMar 2026
Price in / out$0.21 / $0.84
Context197K
Core tests taken1 of 9
Claude Opus 4.7 leads MiniMax-M2.7 by 34 points, ahead on 1 of 1 shared core tests, at 27× the price.
Test by testthird-party results
Claude Opus 4.7
MiniMax-M2.7
Every result as a table
| Test | Claude Opus 4.7 | MiniMax-M2.7 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Vals AI) | 90.1% | 86.6% |
| MMLU Pro | 89.9% | 80.4% |
| Coding | ||
| Code Migration | 43.9% | 8.7% |
| LiveCodeBench | 85.1% | 79.9% |
| ProgramBench | 0.0% | 0.0% |
| SWE-bench | 82.0% | 73.8% |
| Vibe Code Bench v1.1 | 71.0% | 11.9% |
| WeirdML | 76.4% | 37.0% |
| Professional work | ||
| Finance Agent (v2) | 51.5% | 27.9% |
| CorpFin v2 | 66.1% | 61.2% |
| EMB | 63.7% | 28.4% |
| LegalBench | 85.3% | 84.0% |
| Legal Research Bench | 38.5% | 10.6% |
| MedCode | 54.9% | 34.4% |
| MedScribe | 83.0% | 79.9% |
| Tax Agent Bench | 23.7% | 6.9% |
| TaxEval v2 | 75.3% | 66.6% |
Tests at least two of these models have taken. Bold rows are the core tests.