Model comparison
MiniMax-M2.5 vs Mixtral 8x7B
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 27.1 on the Noometry Index.
Last verified . 18 shared benchmarks.
MiniMax
Mistral AISummary
- They share 18 benchmarks with published results for both. MiniMax-M2.5 scores higher in 7 categories and Mixtral 8x7B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2.5 leads 39.2 to 11.0.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.70 / $0.70 for Mixtral 8x7B.
- MiniMax-M2.5 accepts more context: 205K tokens versus 32K.
Side by side
| MiniMax-M2.5 | Mixtral 8x7B | |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 38.3 | 27.1 |
| Released | 2026-02-12 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 205K | 32K |
| Max output | 131K | 32K |
| Input $ / M tokens | $0.30 | $0.70 |
| Output $ / M tokens | $1.20 | $0.70 |
| Results tracked | 33 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
MiniMax-M2.5: 48.1 (#58), Mixtral 8x7B: 32.8 (#269)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | 1381 | 1126 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| ALE-Bench | 618.17 | — |
| HumanEval+ | — | 39.6% |
| MBPP+ | — | 49.7% |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.4 (#77), Mixtral 8x7B: —
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning Too close to call
MiniMax-M2.5: 17.5 (#292), Mixtral 8x7B: 18.2 (#285)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1115 |
| Epoch Capabilities Index | 146.68 | 118.47 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| DTBench | — | 49.6% |
| Adversarial NLI | — | 55.2% |
| ForecastBench | — | 56.3 |
| HellaSwag | — | 86.7% |
| PIQA | — | 83.6% |
| WinoGrande | — | 77.2% |
Math MiniMax-M2.5 leads
MiniMax-M2.5: 26.9 (#253), Mixtral 8x7B: 18.8 (#289)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Math | 1378 | 1147 |
| ProofBench | 4% | — |
| Omni-MATH | — | 10.5% |
| MATH Level 5 | — | 10% |
| GSM8K | — | 74.4% |
Knowledge MiniMax-M2.5 leads
MiniMax-M2.5: 39.2 (#135), Mixtral 8x7B: 11.0 (#301)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Expert | 1379 | 1088 |
| GPQA Diamond | — | 30.6% |
| MMLU-Pro | — | 33.5% |
| Vectara Hallucination Rate | 9.1% | — |
| GPQA (HELM) | — | 29.6% |
| ARC (AI2) Challenge | — | 87.3% |
| MMLU | — | 70.6% |
| OpenBookQA | — | 85.8% |
| TriviaQA | — | 82.2% |
Multilingual MiniMax-M2.5 leads
MiniMax-M2.5: 47.1 (#152), Mixtral 8x7B: 29.6 (#266)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 1338 | 1077 |
| LMArena Chinese | 1393 | 1055 |
| LMArena French | 1362 | 1166 |
| LMArena German | 1362 | 1114 |
| LMArena Japanese | 1171 | 931 |
| LMArena Korean | 1232 | 968 |
| LMArena Russian | 1358 | 1090 |
| LMArena Spanish | 1354 | 1111 |
Instruction Following MiniMax-M2.5 leads
MiniMax-M2.5: 71.5 (#148), Mixtral 8x7B: 51.0 (#297)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Instruction Following | 1353 | 1109 |
| IFEval | — | 57.5% |
Long Context MiniMax-M2.5 leads
MiniMax-M2.5: 37.5 (#216), Mixtral 8x7B: 33.4 (#260)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | 1366 | 1103 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
Writing & Preference MiniMax-M2.5 leads
MiniMax-M2.5: 53.9 (#153), Mixtral 8x7B: 34.2 (#270)
| Benchmark | MiniMax-M2.5 | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1359 | 1132 |
| LMArena Creative Writing | 1331 | 1109 |
| LMArena Multi-Turn | 1364 | 1115 |
| EQ-Bench Creative Writing | 1361 | — |
| WildBench | — | 67.3% |
Frequently asked questions
Is MiniMax-M2.5 better than Mixtral 8x7B?
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 27.1 on the Noometry Index.
Which is cheaper, MiniMax-M2.5 or Mixtral 8x7B?
MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Mixtral 8x7B lists at $0.70 and $0.70.
Is MiniMax-M2.5 or Mixtral 8x7B better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 32.8 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 32K.
How many benchmarks do MiniMax-M2.5 and Mixtral 8x7B share?
18 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Mixtral 8x7B has 38.