Model comparison
MiniMax-M2.7 vs Qwen3-4B
MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 31.5 on the Noometry Index.
Last verified . 1 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. MiniMax-M2.7 scores higher in 2 categories and Qwen3-4B in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2.7 leads 37.1 to 33.0.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 12.9% for MiniMax-M2.7 and 5.7% for Qwen3-4B.
Side by side
| MiniMax-M2.7 | Qwen3-4B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 37.4 | 31.5 |
| Released | 2026-03-18 | 2025-04-29 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 30 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
MiniMax-M2.7: 41.4 (#115), Qwen3-4B: —
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| LMArena WebDev | 1397 | — |
| SciCode | 47% | — |
| WeirdML | 37% | — |
| LMArena Coding | 1454 | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use Qwen3-4B leads
MiniMax-M2.7: 24.9 (#103), Qwen3-4B: 27.3 (#94)
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| Berkeley Function Calling Leaderboard | — | 35.7% |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Too close to call
MiniMax-M2.7: 19.0 (#248), Qwen3-4B: 18.5 (#264)
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| NYT Connections (extended) | 24.7% | — |
| CritPt | 0.6% | — |
| Chess Puzzles | — | 4% |
| Thematic Generalization | 39.3% | — |
| LMArena Hard Prompts | 1422 | — |
| Epoch Capabilities Index | 145.84 | — |
Math Qwen3-4B leads
MiniMax-M2.7: 24.9 (#254), Qwen3-4B: 28.5 (#228)
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| ProofBench | 3% | — |
| LMArena Math | 1420 | — |
Knowledge MiniMax-M2.7 leads
MiniMax-M2.7: 37.1 (#151), Qwen3-4B: 33.0 (#198)
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| Vectara Hallucination Rate | 12.9% | 5.7% |
| GPQA Diamond | — | 52.3% |
| LMArena Expert | 1443 | — |
Multilingual Not comparable
MiniMax-M2.7: 50.5 (#120), Qwen3-4B: —
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1382 | — |
| LMArena Chinese | 1439 | — |
| LMArena French | 1422 | — |
| LMArena German | 1398 | — |
| LMArena Japanese | 1261 | — |
| LMArena Korean | 1313 | — |
| LMArena Russian | 1383 | — |
| LMArena Spanish | 1403 | — |
Instruction Following Not comparable
MiniMax-M2.7: 74.2 (#101), Qwen3-4B: —
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1404 | — |
Long Context Not comparable
MiniMax-M2.7: 43.3 (#94), Qwen3-4B: —
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| LMArena Longer Query | 1419 | — |
Writing & Preference Not comparable
MiniMax-M2.7: 59.1 (#107), Qwen3-4B: —
| Benchmark | MiniMax-M2.7 | Qwen3-4B |
|---|---|---|
| LMArena Text | 1404 | — |
| LMArena Creative Writing | 1354 | — |
| LMArena Multi-Turn | 1413 | — |
Frequently asked questions
Is MiniMax-M2.7 better than Qwen3-4B?
MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 31.5 on the Noometry Index.
How many benchmarks do MiniMax-M2.7 and Qwen3-4B share?
1 benchmark has published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Qwen3-4B has 6.