Model comparison
MiniMax-M3 vs Qwen3-30B-A3B
MiniMax-M3 is the stronger model overall, scoring 43.5 to 38.5 on the Noometry Index. Qwen3-30B-A3B costs 2.4× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. MiniMax-M3 scores higher in 8 categories and Qwen3-30B-A3B in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 41.7.
- The biggest single-benchmark swing is GPQA Diamond: 90.9% for MiniMax-M3 and 70.1% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 41K.
Side by side
| MiniMax-M3 | Qwen3-30B-A3B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 43.5 | 38.5 |
| Released | 2026-06-01 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 1M | 41K |
| Max output | 512K | 16K |
| Input $ / M tokens | $0.30 | $0.12 |
| Output $ / M tokens | $1.20 | $0.50 |
| Results tracked | 41 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.5 (#111), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| SciCode | 47.1% | 33.3% |
| LMArena Coding | 1470 | 1416 |
| FrontierCode | 14.7% | — |
| LMArena WebDev | 1482 | — |
| WeirdML | — | 29.8% |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use Qwen3-30B-A3B leads
MiniMax-M3: 22.5 (#121), Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| APEX-Agents | 37.7% | — |
| Berkeley Function Calling Leaderboard | — | 41.4% |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning MiniMax-M3 leads
MiniMax-M3: 29.3 (#83), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| CritPt | 3.7% | 0.3% |
| Chess Puzzles | 14% | 8% |
| LMArena Hard Prompts | 1447 | 1398 |
| DTBench | 78.9% | 69.3% |
| LMCA | 33.7% | 22.4% |
| Epoch Capabilities Index | 147 | 139.63 |
| SimpleBench | 45.8% | — |
| Kagi LLM Benchmark | — | 54.9% |
| NYT Connections (extended) | 65.1% | — |
| Mystery Game Puzzles | 8% | — |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |
Math MiniMax-M3 leads
MiniMax-M3: 39.1 (#83), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 70.3% |
| LMArena Math | 1432 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| ProofBench | 18% | — |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#32), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 90.9% | 70.1% |
| LMArena Expert | 1462 | 1395 |
| Confabulations | — | 12.3% |
Multimodal Not comparable
MiniMax-M3: 39.5 (#45), Qwen3-30B-A3B: —
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Vision | 1254 | — |
| LMArena Document | 1435 | — |
Multilingual MiniMax-M3 leads
MiniMax-M3: 53.2 (#70), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1420 | 1372 |
| LMArena Chinese | 1465 | 1434 |
| LMArena French | 1445 | 1419 |
| LMArena German | 1424 | 1379 |
| LMArena Japanese | 1381 | 1338 |
| LMArena Korean | 1372 | 1333 |
| LMArena Russian | 1430 | 1369 |
| LMArena Spanish | 1431 | 1403 |
Instruction Following MiniMax-M3 leads
MiniMax-M3: 75.7 (#56), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1364 |
Long Context MiniMax-M3 leads
MiniMax-M3: 44.2 (#68), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1445 | 1379 |
| Fiction.LiveBench | — | 40.6% |
Writing & Preference MiniMax-M3 leads
MiniMax-M3: 62.3 (#79), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | MiniMax-M3 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1433 | 1384 |
| LMArena Creative Writing | 1405 | 1318 |
| LMArena Multi-Turn | 1441 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
| EQ-Bench 4 | 1150 | — |
Frequently asked questions
Is MiniMax-M3 better than Qwen3-30B-A3B?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 38.5 on the Noometry Index. Qwen3-30B-A3B costs 2.4× 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 Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is MiniMax-M3 or Qwen3-30B-A3B better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 37.0 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 41K.
How many benchmarks do MiniMax-M3 and Qwen3-30B-A3B share?
25 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Qwen3-30B-A3B has 32.