Model comparison
MiniMax-M2 vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 36.9 on the Noometry Index. MiniMax-M2 costs 2.1× less per token, which makes it the better buy when Qwen3.8 27B'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 scores higher in 0 categories and Qwen3.8 27B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 31.5 to 18.6.
- The biggest single-benchmark swing is NYT Connections (extended): 14.8% for MiniMax-M2 and 54.5% for Qwen3.8 27B.
- MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.99 / $1.49 for Qwen3.8 27B.
- Qwen3.8 27B accepts more context: 262K tokens versus 205K.
Side by side
| MiniMax-M2 | Qwen3.8 27B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 36.9 | 45.3 |
| Released | 2025-10-27 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 205K | 262K |
| Max output | 131K | 33K |
| Input $ / M tokens | $0.30 | $0.99 |
| Output $ / M tokens | $1.20 | $1.49 |
| Results tracked | 21 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
MiniMax-M2: 39.1 (#151), Qwen3.8 27B: 50.3 (#42)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1297 | 1592 |
| LMArena Coding | 1370 | 1487 |
| SWE-bench Verified (bash only) | 61% | — |
| SciCode | — | 46.6% |
Agentic & Tool Use Not comparable
MiniMax-M2: 24.9 (#102), Qwen3.8 27B: —
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| Terminal-Bench | 30% | — |
| Vending-Bench 2 | 160.6 | — |
Reasoning Qwen3.8 27B leads
MiniMax-M2: 18.6 (#256), Qwen3.8 27B: 31.5 (#71)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 14.8% | 54.5% |
| LMArena Hard Prompts | 1357 | 1459 |
| Kagi LLM Benchmark | 57.8% | — |
| CritPt | — | 5.4% |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
| Epoch Capabilities Index | — | 149.38 |
Math Qwen3.8 27B leads
MiniMax-M2: 34.6 (#158), Qwen3.8 27B: 35.7 (#139)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1350 | 1453 |
| ProofBench | — | 16% |
Knowledge Qwen3.8 27B leads
MiniMax-M2: 36.3 (#160), Qwen3.8 27B: 40.9 (#105)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1336 | 1482 |
Multimodal Not comparable
MiniMax-M2: —, Qwen3.8 27B: 40.5 (#36)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1270 |
Multilingual Qwen3.8 27B leads
MiniMax-M2: 45.6 (#165), Qwen3.8 27B: 54.0 (#54)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1314 | 1431 |
| LMArena Chinese | 1367 | 1504 |
| LMArena French | 1336 | 1468 |
| LMArena German | 1366 | 1435 |
| LMArena Russian | 1332 | 1421 |
| LMArena Spanish | 1326 | 1444 |
| LMArena Japanese | — | 1392 |
| LMArena Korean | — | 1389 |
Instruction Following Qwen3.8 27B leads
MiniMax-M2: 70.4 (#160), Qwen3.8 27B: 75.8 (#54)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1329 | 1436 |
Long Context Qwen3.8 27B leads
MiniMax-M2: 40.5 (#148), Qwen3.8 27B: 44.3 (#66)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1332 | 1448 |
Writing & Preference Qwen3.8 27B leads
MiniMax-M2: 53.3 (#157), Qwen3.8 27B: 65.9 (#41)
| Benchmark | MiniMax-M2 | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1340 | 1441 |
| LMArena Creative Writing | 1288 | 1382 |
| LMArena Multi-Turn | 1362 | 1443 |
| EQ-Bench Creative Writing | — | 1671 |
Frequently asked questions
Is MiniMax-M2 better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 36.9 on the Noometry Index. MiniMax-M2 costs 2.1× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2 or Qwen3.8 27B?
MiniMax-M2 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.8 27B lists at $0.99 and $1.49.
Is MiniMax-M2 or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.3 versus 39.1 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 27B does, with 262K tokens against 205K.
How many benchmarks do MiniMax-M2 and Qwen3.8 27B share?
17 benchmarks have published results for both models. MiniMax-M2 has 21 scored results on Noometry and Qwen3.8 27B has 28.