Up to four models
Mistral Large 4 vs MiniMax-M3
On the index (GPT-4 = 100), Mistral Large 4 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.
148132–165
WeightsOpen
ReleasedOct 2026
Price in / out$0.70 / $2.1
Context1.05M
Core tests taken3 of 9
147131–164
WeightsOpen
ReleasedJun 2026
Price in / out$0.30 / $1.2
Context1.05M
Core tests taken4 of 9
Mistral Large 4 and MiniMax-M3 are too close to call (1 point apart, within the uncertainty), ahead on 1 of 2 shared core tests, at 2.0× the price.
Test by testthird-party results
Every result as a table
| Test | Mistral Large 4 | MiniMax-M3 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 88.1% | 90.9% |
| Maths | ||
| AIME-style maths (OTIS mock) | 98.3% | 71.1% |
| Chess puzzles | 33.0% | 17.0% |
| Surface Evolver Bench | 28.1% | 55.0% |
| Agents | ||
| APEX-Agents | 51.6% | 37.7% |
| DTBench | 88.2% | 78.9% |
| LMCA | 35.8% | 33.7% |
| Mystery game puzzles | 17.0% | 8.0% |
Tests at least two of these models have taken. Bold rows are the core tests.