Model comparison
MiniMax-M3 vs Qwen3.5 35B-A3B
MiniMax-M3 is the stronger model overall, scoring 43.8 to 42.0 on the Noometry Index.
Last verified . 26 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. MiniMax-M3 scores higher in 8 categories and Qwen3.5 35B-A3B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 47.8.
- The biggest single-benchmark swing is SciCode: 47.1% for MiniMax-M3 and 29.3% for Qwen3.5 35B-A3B.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.25 / $2 for Qwen3.5 35B-A3B.
- MiniMax-M3 accepts more context: 1M tokens versus 262K.
Side by side
| MiniMax-M3 | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 43.8 | 42.0 |
| Released | 2026-06-01 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 512K | 66K |
| Input $ / M tokens | $0.30 | $0.25 |
| Output $ / M tokens | $1.20 | $2 |
| Results tracked | 41 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.8 (#118), Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena WebDev | 1482 | 1254 |
| SciCode | 47.1% | 29.3% |
| LMArena Coding | 1469 | 1410 |
| FrontierCode | 14.7% | — |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use Not comparable
MiniMax-M3: 22.6 (#130), Qwen3.5 35B-A3B: —
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| APEX-Agents | 37.7% | — |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning MiniMax-M3 leads
MiniMax-M3: 30.1 (#87), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| CritPt | 3.7% | 0.6% |
| Chess Puzzles | 14% | 10% |
| LMArena Hard Prompts | 1447 | 1400 |
| DTBench | 78.9% | 80% |
| LMCA | 33.7% | 29.5% |
| Epoch Capabilities Index | 146.95 | 142.52 |
| SimpleBench | 45.8% | — |
| NYT Connections (extended) | 65.1% | — |
| Mystery Game Puzzles | 8% | — |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |
Math Too close to call
MiniMax-M3: 40.0 (#95), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 70% |
| LMArena Math | 1429 | 1404 |
| MathArena Final-Answer Competitions | — | 56% |
| ProofBench | 18% | — |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#35), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| GPQA Diamond | 90.9% | 83.5% |
| LMArena Expert | 1461 | 1408 |
| Vectara Hallucination Rate | — | 10.5% |
Multimodal Not comparable
MiniMax-M3: 40.2 (#51), Qwen3.5 35B-A3B: —
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Vision | 1253 | — |
| LMArena Document | 1435 | — |
Multilingual MiniMax-M3 leads
MiniMax-M3: 53.0 (#75), Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | 1420 | 1378 |
| LMArena Chinese | 1463 | 1457 |
| LMArena French | 1447 | 1412 |
| LMArena German | 1426 | 1367 |
| LMArena Japanese | 1381 | 1325 |
| LMArena Korean | 1372 | 1356 |
| LMArena Russian | 1428 | 1376 |
| LMArena Spanish | 1432 | 1392 |
Instruction Following MiniMax-M3 leads
MiniMax-M3: 75.5 (#62), Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1379 |
Long Context MiniMax-M3 leads
MiniMax-M3: 44.2 (#72), Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1445 | 1389 |
Writing & Preference MiniMax-M3 leads
MiniMax-M3: 62.1 (#83), Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | MiniMax-M3 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | 1433 | 1395 |
| LMArena Creative Writing | 1404 | 1346 |
| LMArena Multi-Turn | 1442 | 1390 |
| EQ-Bench 4 | 1150 | — |
Frequently asked questions
Is MiniMax-M3 better than Qwen3.5 35B-A3B?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 42.0 on the Noometry Index.
Which is cheaper, MiniMax-M3 or Qwen3.5 35B-A3B?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.5 35B-A3B lists at $0.25 and $2.
Is MiniMax-M3 or Qwen3.5 35B-A3B better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 262K.
How many benchmarks do MiniMax-M3 and Qwen3.5 35B-A3B share?
26 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Qwen3.5 35B-A3B has 28.