Model comparison
MiniMax-M3 vs Qwen3.7 Max
Qwen3.7 Max is the stronger model overall, scoring 51.5 to 43.8 on the Noometry Index. MiniMax-M3 costs 7.1× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 28 benchmarks with published results for both. MiniMax-M3 scores higher in 1 category and Qwen3.7 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.7 Max leads 62.4 to 40.0.
- The biggest single-benchmark swing is SimpleBench: 45.8% for MiniMax-M3 and 70.4% for Qwen3.7 Max.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M3 | Qwen3.7 Max | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 43.8 | 51.5 |
| Released | 2026-06-01 | 2026-05-19 |
| Weights | Open | Proprietary |
| Context window | 1M | 1M |
| Max output | 512K | 131K |
| Input $ / M tokens | $0.30 | $2.50 |
| Output $ / M tokens | $1.20 | $7.50 |
| Results tracked | 41 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.7 Max leads
MiniMax-M3: 41.8 (#118), Qwen3.7 Max: 50.4 (#45)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| LMArena WebDev | 1482 | 1515 |
| SciCode | 47.1% | 48.8% |
| LMArena Coding | 1469 | 1498 |
| ALE-Bench | 640.02 | 1,189 |
| SWE-bench Verified | — | 77.3% |
| FrontierCode | 14.7% | — |
Agentic & Tool Use Too close to call
MiniMax-M3: 22.6 (#130), Qwen3.7 Max: 22.1 (#135)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| GBAEval | 0.9% | 0.4% |
| APEX-Agents | 37.7% | — |
| OSWorld 2.0 | 4.6% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning Qwen3.7 Max leads
MiniMax-M3: 30.1 (#87), Qwen3.7 Max: 49.2 (#38)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| SimpleBench | 45.8% | 70.4% |
| NYT Connections (extended) | 65.1% | 85.1% |
| CritPt | 3.7% | 13.4% |
| Chess Puzzles | 14% | 19% |
| LMArena Hard Prompts | 1447 | 1483 |
| Mystery Game Puzzles | 8% | 32% |
| DTBench | 78.9% | 92.3% |
| LMCA | 33.7% | 44% |
| Epoch Capabilities Index | 146.95 | 153.68 |
| EBR-Bench | — | 9.5% |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |
Math Qwen3.7 Max leads
MiniMax-M3: 40.0 (#95), Qwen3.7 Max: 62.4 (#32)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 95.6% |
| ProofBench | 18% | 26% |
| LMArena Math | 1429 | 1490 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 34.1% |
Knowledge Qwen3.7 Max leads
MiniMax-M3: 58.4 (#35), Qwen3.7 Max: 61.6 (#28)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | 90.9% | 90.9% |
| LMArena Expert | 1461 | 1488 |
| SimpleQA Verified | — | 55.8% |
Multimodal Not comparable
MiniMax-M3: 40.2 (#51), Qwen3.7 Max: —
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| LMArena Vision | 1253 | — |
| LMArena Document | 1435 | — |
Multilingual Qwen3.7 Max leads
MiniMax-M3: 53.0 (#75), Qwen3.7 Max: 56.9 (#15)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1420 | 1474 |
| LMArena Chinese | 1463 | 1530 |
| LMArena Russian | 1428 | 1484 |
| LMArena French | 1447 | — |
| LMArena German | 1426 | — |
| LMArena Japanese | 1381 | — |
| LMArena Korean | 1372 | — |
| LMArena Spanish | 1432 | — |
Instruction Following Qwen3.7 Max leads
MiniMax-M3: 75.5 (#62), Qwen3.7 Max: 76.7 (#38)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1433 | 1460 |
Long Context Qwen3.7 Max leads
MiniMax-M3: 44.2 (#72), Qwen3.7 Max: 45.4 (#40)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1445 | 1482 |
Writing & Preference Qwen3.7 Max leads
MiniMax-M3: 62.1 (#83), Qwen3.7 Max: 65.0 (#54)
| Benchmark | MiniMax-M3 | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1433 | 1476 |
| LMArena Creative Writing | 1404 | 1449 |
| EQ-Bench 4 | 1150 | 1110 |
| LMArena Multi-Turn | 1442 | 1481 |
Frequently asked questions
Is MiniMax-M3 better than Qwen3.7 Max?
Qwen3.7 Max is the stronger model overall, scoring 51.5 to 43.8 on the Noometry Index. MiniMax-M3 costs 7.1× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M3 or Qwen3.7 Max?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.
Is MiniMax-M3 or Qwen3.7 Max better for coding?
Qwen3.7 Max scores higher on coding benchmarks: 50.4 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do MiniMax-M3 and Qwen3.7 Max share?
28 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Qwen3.7 Max has 33.