Model comparison
MiniMax-M2.5 vs Qwen3-4B
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 31.9 on the Noometry Index.
Last verified . 1 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. MiniMax-M2.5 scores higher in 2 categories and Qwen3-4B in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2.5 leads 39.2 to 33.0.
Side by side
| MiniMax-M2.5 | Qwen3-4B | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 38.3 | 31.9 |
| Released | 2026-02-12 | 2025-04-29 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 33 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
MiniMax-M2.5: 48.1 (#58), Qwen3-4B: —
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| LMArena Coding | 1381 | — |
| ALE-Bench | 618.17 | — |
Agentic & Tool Use MiniMax-M2.5 leads
MiniMax-M2.5: 30.4 (#77), Qwen3-4B: 27.6 (#100)
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Berkeley Function Calling Leaderboard | — | 35.7% |
| Vending-Bench 2 | -23.16 | — |
Reasoning Qwen3-4B leads
MiniMax-M2.5: 17.5 (#292), Qwen3-4B: 19.2 (#268)
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1372 | — |
| Epoch Capabilities Index | 146.68 | — |
Math Qwen3-4B leads
MiniMax-M2.5: 26.9 (#253), Qwen3-4B: 29.7 (#240)
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| ProofBench | 4% | — |
| LMArena Math | 1378 | — |
Knowledge MiniMax-M2.5 leads
MiniMax-M2.5: 39.2 (#135), Qwen3-4B: 33.0 (#208)
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| Vectara Hallucination Rate | 9.1% | 5.7% |
| GPQA Diamond | — | 52.3% |
| LMArena Expert | 1379 | — |
Multilingual Not comparable
MiniMax-M2.5: 47.1 (#152), Qwen3-4B: —
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1338 | — |
| LMArena Chinese | 1393 | — |
| LMArena French | 1362 | — |
| LMArena German | 1362 | — |
| LMArena Japanese | 1171 | — |
| LMArena Korean | 1232 | — |
| LMArena Russian | 1358 | — |
| LMArena Spanish | 1354 | — |
Instruction Following Not comparable
MiniMax-M2.5: 71.5 (#148), Qwen3-4B: —
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1353 | — |
Long Context Not comparable
MiniMax-M2.5: 37.5 (#216), Qwen3-4B: —
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
| LMArena Longer Query | 1366 | — |
Writing & Preference Not comparable
MiniMax-M2.5: 53.9 (#153), Qwen3-4B: —
| Benchmark | MiniMax-M2.5 | Qwen3-4B |
|---|---|---|
| LMArena Text | 1359 | — |
| LMArena Creative Writing | 1331 | — |
| EQ-Bench Creative Writing | 1361 | — |
| LMArena Multi-Turn | 1364 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Qwen3-4B?
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 31.9 on the Noometry Index.
How many benchmarks do MiniMax-M2.5 and Qwen3-4B share?
1 benchmark has published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Qwen3-4B has 6.