Up to four models
Qwen3.7-Plus vs MiniMax-M3
On the index (GPT-4 = 100), Qwen3.7-Plus scores 148 and MiniMax-M3 147. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
148131–165
WeightsClosed
ReleasedJun 2026
Price in / out$0.32 / $1.28
Context1M
Core tests taken3 of 9
147131–164
WeightsOpen
ReleasedJun 2026
Price in / out$0.30 / $1.2
Context1.05M
Core tests taken4 of 9
Qwen3.7-Plus and MiniMax-M3 are too close to call (1 point apart, within the uncertainty), ahead on 0 of 2 shared core tests, at a similar price.
Test by testthird-party results
Every result as a table
| Test | Qwen3.7-Plus | MiniMax-M3 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 87.9% | 90.9% |
| Maths | ||
| AIME-style maths (OTIS mock) | 93.3% | 71.1% |
| Coding | ||
| Code Migration | 12.9% | 19.9% |
| Vibe Code Bench v1.1 | 46.4% | 47.6% |
| Agents | ||
| SkillsBench | 54.3% | 51.5% |
| DTBench | 84.0% | 78.9% |
| LMCA | 37.6% | 33.7% |
| Mystery game puzzles | 17.0% | 8.0% |
| Professional work | ||
| Finance Agent (v2) | 38.2% | 48.3% |
| EMB | 49.3% | 47.8% |
| Legal Research Bench | 16.4% | 29.8% |
| MortgageTax | 66.2% | 68.4% |
| SAGE | 89.1% | 85.5% |
| Tax Agent Bench | 10.5% | 20.6% |
Tests at least two of these models have taken. Bold rows are the core tests.