Model comparison
MiniMax-M2.1 vs Mistral Medium
MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 36.3 on the Noometry Index.
Last verified . 19 shared benchmarks.
MiniMax
Mistral AISummary
- They share 19 benchmarks with published results for both. MiniMax-M2.1 scores higher in 5 categories and Mistral Medium in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2.1 leads 38.3 to 25.0.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 11.8% for MiniMax-M2.1 and 22.7% for Mistral Medium.
- MiniMax-M2.1 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 205K.
Side by side
| MiniMax-M2.1 | Mistral Medium | |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 38.9 | 36.3 |
| Released | 2025-12-23 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 205K | 262K |
| Max output | 131K | 262K |
| Input $ / M tokens | $0.30 | $1.50 |
| Output $ / M tokens | $1.20 | $7.50 |
| Results tracked | 22 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.1 leads
MiniMax-M2.1: 40.4 (#143), Mistral Medium: 34.2 (#243)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Coding | 1421 | 1434 |
| ALE-Bench | 623.83 | 763.98 |
| FrontierCode | — | 8% |
| LMArena WebDev | 1384 | — |
| SciCode | — | 40.2% |
| WeirdML | — | 43.7% |
Agentic & Tool Use Too close to call
MiniMax-M2.1: 27.9 (#98), Mistral Medium: 28.3 (#90)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| Terminal-Bench | 36.6% | — |
| Berkeley Function Calling Leaderboard | — | 37.7% |
Reasoning Mistral Medium leads
MiniMax-M2.1: 16.6 (#302), Mistral Medium: 24.0 (#167)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Hard Prompts | 1411 | 1426 |
| Kagi LLM Benchmark | — | 50% |
| NYT Connections (extended) | 11.2% | — |
| CritPt | — | 0% |
| DTBench | — | 75.5% |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
Math MiniMax-M2.1 leads
MiniMax-M2.1: 38.3 (#138), Mistral Medium: 28.1 (#245)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Math | 1397 | 1408 |
| OTIS Mock AIME 2024-2025 | — | 32.2% |
| ProofBench | — | 9% |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge MiniMax-M2.1 leads
MiniMax-M2.1: 38.3 (#147), Mistral Medium: 25.0 (#265)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| Vectara Hallucination Rate | 11.8% | 22.7% |
| LMArena Expert | 1431 | 1408 |
| GPQA Diamond | — | 59.5% |
| Humanity's Last Exam | — | 4.5% |
Multimodal Not comparable
MiniMax-M2.1: —, Mistral Medium: 35.3 (#88)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Mistral Medium leads
MiniMax-M2.1: 50.0 (#128), Mistral Medium: 52.1 (#91)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1378 | 1408 |
| LMArena Chinese | 1430 | 1447 |
| LMArena French | 1404 | 1459 |
| LMArena German | 1381 | 1432 |
| LMArena Japanese | 1287 | 1378 |
| LMArena Korean | 1298 | 1380 |
| LMArena Russian | 1387 | 1411 |
| LMArena Spanish | 1397 | 1433 |
Instruction Following Too close to call
MiniMax-M2.1: 73.8 (#112), Mistral Medium: 73.7 (#116)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1400 | 1398 |
Long Context Too close to call
MiniMax-M2.1: 43.2 (#101), Mistral Medium: 42.9 (#114)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1416 | 1406 |
Writing & Preference Mistral Medium leads
MiniMax-M2.1: 58.3 (#120), Mistral Medium: 60.0 (#103)
| Benchmark | MiniMax-M2.1 | Mistral Medium |
|---|---|---|
| LMArena Text | 1392 | 1424 |
| LMArena Creative Writing | 1361 | 1391 |
| LMArena Multi-Turn | 1396 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
Frequently asked questions
Is MiniMax-M2.1 better than Mistral Medium?
MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 36.3 on the Noometry Index.
Which is cheaper, MiniMax-M2.1 or Mistral Medium?
MiniMax-M2.1 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is MiniMax-M2.1 or Mistral Medium better for coding?
MiniMax-M2.1 scores higher on coding benchmarks: 40.4 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 205K.
How many benchmarks do MiniMax-M2.1 and Mistral Medium share?
19 benchmarks have published results for both models. MiniMax-M2.1 has 22 scored results on Noometry and Mistral Medium has 36.