Model comparison
MiniMax-M2.7 vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 37.7 on the Noometry Index.
Last verified . 23 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. MiniMax-M2.7 scores higher in 0 categories and Qwen3.8 27B in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 19.7.
- The biggest single-benchmark swing is NYT Connections (extended): 24.7% for MiniMax-M2.7 and 54.5% for Qwen3.8 27B.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.04 / $2.30 for Qwen3.8 27B.
- Qwen3.8 27B accepts more context: 262K tokens versus 205K.
Side by side
| MiniMax-M2.7 | Qwen3.8 27B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 37.7 | 46.0 |
| Released | 2026-03-18 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 205K | 262K |
| Max output | 131K | 33K |
| Input $ / M tokens | $0.30 | $0.04 |
| Output $ / M tokens | $1.20 | $2.30 |
| Results tracked | 30 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
MiniMax-M2.7: 41.8 (#120), Qwen3.8 27B: 50.5 (#44)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1398 | 1593 |
| SciCode | 47% | 46.6% |
| LMArena Coding | 1454 | 1482 |
| WeirdML | 37% | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use Qwen3.8 27B leads
MiniMax-M2.7: 25.1 (#111), Qwen3.8 27B: 32.9 (#57)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| APEX-Agents | — | 47.5% |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Qwen3.8 27B leads
MiniMax-M2.7: 19.7 (#253), Qwen3.8 27B: 41.0 (#54)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 24.7% | 54.5% |
| CritPt | 0.6% | 5.4% |
| LMArena Hard Prompts | 1422 | 1460 |
| Epoch Capabilities Index | 145.85 | 149.38 |
| ARC-AGI-2 | — | 42.4% |
| ARC-AGI-1 | — | 87.5% |
| Thematic Generalization | 39.3% | — |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
Math Qwen3.8 27B leads
MiniMax-M2.7: 25.9 (#263), Qwen3.8 27B: 37.1 (#161)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| ProofBench | 3% | 16% |
| LMArena Math | 1420 | 1456 |
Knowledge Qwen3.8 27B leads
MiniMax-M2.7: 37.7 (#152), Qwen3.8 27B: 41.6 (#109)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1444 | 1482 |
| Vectara Hallucination Rate | 12.9% | — |
Multimodal Not comparable
MiniMax-M2.7: —, Qwen3.8 27B: 41.3 (#37)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1271 |
Multilingual Qwen3.8 27B leads
MiniMax-M2.7: 50.3 (#123), Qwen3.8 27B: 53.7 (#60)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1382 | 1430 |
| LMArena Chinese | 1441 | 1504 |
| LMArena French | 1421 | 1465 |
| LMArena German | 1398 | 1438 |
| LMArena Japanese | 1262 | 1384 |
| LMArena Korean | 1313 | 1393 |
| LMArena Russian | 1383 | 1415 |
| LMArena Spanish | 1403 | 1448 |
Instruction Following Qwen3.8 27B leads
MiniMax-M2.7: 74.1 (#103), Qwen3.8 27B: 75.8 (#53)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1405 | 1439 |
Long Context Qwen3.8 27B leads
MiniMax-M2.7: 43.3 (#99), Qwen3.8 27B: 44.3 (#70)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1419 | 1450 |
Writing & Preference Qwen3.8 27B leads
MiniMax-M2.7: 58.9 (#112), Qwen3.8 27B: 65.8 (#43)
| Benchmark | MiniMax-M2.7 | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1405 | 1441 |
| LMArena Creative Writing | 1354 | 1384 |
| LMArena Multi-Turn | 1412 | 1441 |
| EQ-Bench Creative Writing | — | 1671 |
Frequently asked questions
Is MiniMax-M2.7 better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 37.7 on the Noometry Index.
Which is cheaper, MiniMax-M2.7 or Qwen3.8 27B?
MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.8 27B lists at $0.04 and $2.30.
Is MiniMax-M2.7 or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 41.8 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-M2.7 and Qwen3.8 27B share?
23 benchmarks have published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Qwen3.8 27B has 31.