Model comparison
MiniMax-M2.7 vs Qwen3 32B
Qwen3 32B is the stronger model overall, scoring 38.8 to 37.4 on the Noometry Index. MiniMax-M2.7 costs 2.3× less per token, which makes it the better buy when Qwen3 32B's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. MiniMax-M2.7 scores higher in 4 categories and Qwen3 32B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 32B leads 37.9 to 24.9.
- The biggest single-benchmark swing is SciCode: 47% for MiniMax-M2.7 and 35.4% for Qwen3 32B.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.70 / $2.80 for Qwen3 32B.
- MiniMax-M2.7 accepts more context: 205K tokens versus 131K.
Side by side
| MiniMax-M2.7 | Qwen3 32B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 37.4 | 38.8 |
| Released | 2026-03-18 | 2025-04 |
| Weights | Open | Open |
| Context window | 205K | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $0.30 | $0.70 |
| Output $ / M tokens | $1.20 | $2.80 |
| Results tracked | 30 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.7 leads
MiniMax-M2.7: 41.4 (#115), Qwen3 32B: 37.3 (#184)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| SciCode | 47% | 35.4% |
| LMArena Coding | 1454 | 1358 |
| Aider Polyglot | — | 40% |
| LMArena WebDev | 1397 | — |
| WeirdML | 37% | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use Qwen3 32B leads
MiniMax-M2.7: 24.9 (#103), Qwen3 32B: 32.3 (#58)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| Berkeley Function Calling Leaderboard | — | 48.7% |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Too close to call
MiniMax-M2.7: 19.0 (#248), Qwen3 32B: 19.6 (#233)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| CritPt | 0.6% | 0.3% |
| LMArena Hard Prompts | 1422 | 1335 |
| Epoch Capabilities Index | 145.84 | 138.51 |
| Kagi LLM Benchmark | — | 54.9% |
| NYT Connections (extended) | 24.7% | — |
| Chess Puzzles | — | 5% |
| Thematic Generalization | 39.3% | — |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
Math Qwen3 32B leads
MiniMax-M2.7: 24.9 (#254), Qwen3 32B: 37.9 (#89)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| LMArena Math | 1420 | 1399 |
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| ProofBench | 3% | — |
Knowledge Qwen3 32B leads
MiniMax-M2.7: 37.1 (#151), Qwen3 32B: 40.0 (#112)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| Vectara Hallucination Rate | 12.9% | 5.9% |
| LMArena Expert | 1443 | 1362 |
| GPQA Diamond | — | 65.7% |
Multilingual MiniMax-M2.7 leads
MiniMax-M2.7: 50.5 (#120), Qwen3 32B: 45.8 (#162)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1382 | 1317 |
| LMArena Chinese | 1439 | 1357 |
| LMArena German | 1398 | 1341 |
| LMArena Russian | 1383 | 1311 |
| LMArena French | 1422 | — |
| LMArena Japanese | 1261 | — |
| LMArena Korean | 1313 | — |
| LMArena Spanish | 1403 | — |
Instruction Following MiniMax-M2.7 leads
MiniMax-M2.7: 74.2 (#101), Qwen3 32B: 69.1 (#174)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1404 | 1305 |
Long Context Too close to call
MiniMax-M2.7: 43.3 (#94), Qwen3 32B: 43.7 (#83)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| LMArena Longer Query | 1419 | 1327 |
| Fiction.LiveBench | — | 74.2% |
Writing & Preference MiniMax-M2.7 leads
MiniMax-M2.7: 59.1 (#107), Qwen3 32B: 53.1 (#158)
| Benchmark | MiniMax-M2.7 | Qwen3 32B |
|---|---|---|
| LMArena Text | 1404 | 1340 |
| LMArena Creative Writing | 1354 | 1297 |
| LMArena Multi-Turn | 1413 | 1331 |
Frequently asked questions
Is MiniMax-M2.7 better than Qwen3 32B?
Qwen3 32B is the stronger model overall, scoring 38.8 to 37.4 on the Noometry Index. MiniMax-M2.7 costs 2.3× less per token, which makes it the better buy when Qwen3 32B's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2.7 or Qwen3 32B?
MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3 32B lists at $0.70 and $2.80.
Is MiniMax-M2.7 or Qwen3 32B better for coding?
MiniMax-M2.7 scores higher on coding benchmarks: 41.4 versus 37.3 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.7 does, with 205K tokens against 131K.
How many benchmarks do MiniMax-M2.7 and Qwen3 32B share?
17 benchmarks have published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Qwen3 32B has 26.