Model comparison
MiniMax-M2 vs Mixtral 8x22B
MiniMax-M2 is the stronger model overall, scoring 37.4 to 27.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
MiniMax
Mistral AISummary
- They share 15 benchmarks with published results for both. MiniMax-M2 scores higher in 8 categories and Mixtral 8x22B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2 leads 37.0 to 15.1.
- MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- MiniMax-M2 accepts more context: 205K tokens versus 64K.
Side by side
| MiniMax-M2 | Mixtral 8x22B | |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 37.4 | 27.1 |
| Released | 2025-10-27 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 205K | 64K |
| Max output | 131K | 64K |
| Input $ / M tokens | $0.30 | $2 |
| Output $ / M tokens | $1.20 | $6 |
| Results tracked | 21 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2 leads
MiniMax-M2: 39.3 (#159), Mixtral 8x22B: 24.2 (#329)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1370 | 1166 |
| SWE-bench Verified (bash only) | 61% | — |
| LMArena WebDev | 1297 | — |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use MiniMax-M2 leads
MiniMax-M2: 25.1 (#109), Mixtral 8x22B: 23.1 (#127)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| Terminal-Bench | 30% | — |
| Cybench | — | 7.5% |
| Vending-Bench 2 | 160.6 | — |
Reasoning Too close to call
MiniMax-M2: 19.4 (#258), Mixtral 8x22B: 19.9 (#248)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1357 | 1150 |
| Kagi LLM Benchmark | 57.8% | — |
| NYT Connections (extended) | 14.8% | — |
| DTBench | — | 55.1% |
| Epoch Capabilities Index | — | 122.03 |
| ForecastBench | — | 56.3 |
Math MiniMax-M2 leads
MiniMax-M2: 37.3 (#160), Mixtral 8x22B: 22.9 (#275)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1352 | 1184 |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
Knowledge MiniMax-M2 leads
MiniMax-M2: 37.0 (#163), Mixtral 8x22B: 15.1 (#293)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Expert | 1337 | 1113 |
| GPQA Diamond | — | 34.1% |
| MMLU-Pro | — | 46% |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multilingual MiniMax-M2 leads
MiniMax-M2: 45.3 (#171), Mixtral 8x22B: 32.8 (#255)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1313 | 1128 |
| LMArena Chinese | 1366 | 1116 |
| LMArena French | 1335 | 1166 |
| LMArena German | 1355 | 1141 |
| LMArena Russian | 1331 | 1158 |
| LMArena Spanish | 1326 | 1151 |
| LMArena Japanese | — | 1037 |
| LMArena Korean | — | 1057 |
Instruction Following MiniMax-M2 leads
MiniMax-M2: 70.2 (#166), Mixtral 8x22B: 57.7 (#266)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1328 | 1147 |
| IFEval | — | 72.4% |
Long Context MiniMax-M2 leads
MiniMax-M2: 40.5 (#153), Mixtral 8x22B: 34.7 (#247)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1331 | 1144 |
Writing & Preference MiniMax-M2 leads
MiniMax-M2: 53.0 (#162), Mixtral 8x22B: 36.9 (#262)
| Benchmark | MiniMax-M2 | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1340 | 1162 |
| LMArena Creative Writing | 1286 | 1141 |
| LMArena Multi-Turn | 1361 | 1130 |
| WildBench | — | 71.1% |
Frequently asked questions
Is MiniMax-M2 better than Mixtral 8x22B?
MiniMax-M2 is the stronger model overall, scoring 37.4 to 27.1 on the Noometry Index.
Which is cheaper, MiniMax-M2 or Mixtral 8x22B?
MiniMax-M2 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is MiniMax-M2 or Mixtral 8x22B better for coding?
MiniMax-M2 scores higher on coding benchmarks: 39.3 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2 does, with 205K tokens against 64K.
How many benchmarks do MiniMax-M2 and Mixtral 8x22B share?
15 benchmarks have published results for both models. MiniMax-M2 has 21 scored results on Noometry and Mixtral 8x22B has 34.