Model comparison
MiniMax-M3 vs Ministral 8B
MiniMax-M3 is the stronger model overall, scoring 43.5 to 27.8 on the Noometry Index. Ministral 8B costs 5.2× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
MiniMax
Mistral AISummary
- They share 14 benchmarks with published results for both. MiniMax-M3 scores higher in 9 categories and Ministral 8B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 12.5.
- The biggest single-benchmark swing is GPQA Diamond: 90.9% for MiniMax-M3 and 27.1% for Ministral 8B.
- Ministral 8B is cheaper at $0.10 / $0.10 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 128K.
Side by side
| MiniMax-M3 | Ministral 8B | |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 43.5 | 27.8 |
| Released | 2026-06-01 | 2024-10-01 |
| Weights | Open | Open |
| Context window | 1M | 128K |
| Max output | 512K | 128K |
| Input $ / M tokens | $0.30 | $0.10 |
| Output $ / M tokens | $1.20 | $0.10 |
| Results tracked | 41 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.5 (#111), Ministral 8B: 34.4 (#226)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Coding | 1470 | 1202 |
| FrontierCode | 14.7% | — |
| LMArena WebDev | 1482 | — |
| SciCode | 47.1% | — |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use MiniMax-M3 leads
MiniMax-M3: 22.5 (#121), Ministral 8B: 16.2 (#140)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| APEX-Agents | 37.7% | — |
| Berkeley Function Calling Leaderboard | — | 11.1% |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning MiniMax-M3 leads
MiniMax-M3: 29.3 (#83), Ministral 8B: 17.7 (#279)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1191 |
| DTBench | 78.9% | 45.7% |
| SimpleBench | 45.8% | — |
| NYT Connections (extended) | 65.1% | — |
| CritPt | 3.7% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| LMCA | 33.7% | — |
| Surface Evolver Bench | 55% | — |
| Epoch Capabilities Index | 147 | — |
| ForecastBench | 61.4 | — |
Math MiniMax-M3 leads
MiniMax-M3: 39.1 (#83), Ministral 8B: 23.7 (#262)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Math | 1432 | 1188 |
| OTIS Mock AIME 2024-2025 | 71.1% | — |
| ProofBench | 18% | — |
| MATH Level 5 | — | 14.9% |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#32), Ministral 8B: 12.5 (#292)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| GPQA Diamond | 90.9% | 27.1% |
| LMArena Expert | 1462 | 1170 |
| Vectara Hallucination Rate | — | 7.4% |
Multimodal Not comparable
MiniMax-M3: 39.5 (#45), Ministral 8B: —
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Vision | 1254 | — |
| LMArena Document | 1435 | — |
Multilingual MiniMax-M3 leads
MiniMax-M3: 53.2 (#70), Ministral 8B: 35.3 (#242)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1420 | 1165 |
| LMArena Chinese | 1465 | 1193 |
| LMArena Russian | 1430 | 1195 |
| LMArena French | 1445 | — |
| LMArena German | 1424 | — |
| LMArena Japanese | 1381 | — |
| LMArena Korean | 1372 | — |
| LMArena Spanish | 1431 | — |
Instruction Following MiniMax-M3 leads
MiniMax-M3: 75.7 (#56), Ministral 8B: 60.7 (#245)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1161 |
Long Context MiniMax-M3 leads
MiniMax-M3: 44.2 (#68), Ministral 8B: 36.7 (#222)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1445 | 1212 |
Writing & Preference MiniMax-M3 leads
MiniMax-M3: 62.3 (#79), Ministral 8B: 39.8 (#241)
| Benchmark | MiniMax-M3 | Ministral 8B |
|---|---|---|
| LMArena Text | 1433 | 1191 |
| LMArena Creative Writing | 1405 | 1175 |
| LMArena Multi-Turn | 1441 | 1166 |
| EQ-Bench 4 | 1150 | — |
Frequently asked questions
Is MiniMax-M3 better than Ministral 8B?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 27.8 on the Noometry Index. Ministral 8B costs 5.2× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M3 or Ministral 8B?
Ministral 8B is cheaper. It lists at $0.10 per million input tokens and $0.10 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is MiniMax-M3 or Ministral 8B better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 128K.
How many benchmarks do MiniMax-M3 and Ministral 8B share?
14 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Ministral 8B has 17.