Model comparison
MiniMax-M2.5 vs Qwen3 Max
Qwen3 Max is the stronger model overall, scoring 43.5 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 4.6× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. MiniMax-M2.5 scores higher in 1 category and Qwen3 Max in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 Max leads 37.6 to 25.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 55.2% for MiniMax-M2.5 and 72.5% for Qwen3 Max.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Qwen3 Max accepts more context: 262K tokens versus 205K.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2.5 | Qwen3 Max | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 38.1 | 43.5 |
| Released | 2026-02-12 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 205K | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $0.30 | $1.20 |
| Output $ / M tokens | $1.20 | $6 |
| Results tracked | 33 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
MiniMax-M2.5: 47.9 (#53), Qwen3 Max: 42.4 (#93)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1380 | 1456 |
| ALE-Bench | 618.17 | 370.45 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1386 | — |
| SWE-bench Multilingual | 68.3% | — |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.2 (#70), Qwen3 Max: —
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | -23.16 | 71.56 |
| Terminal-Bench | 42.7% | — |
Reasoning Qwen3 Max leads
MiniMax-M2.5: 17.6 (#281), Qwen3 Max: 22.0 (#186)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 55.2% | 72.5% |
| NYT Connections (extended) | 16.8% | 30.1% |
| LMArena Hard Prompts | 1372 | 1448 |
| Epoch Capabilities Index | 146.68 | 142.41 |
| ARC-AGI-2 | 4.9% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | — | 4% |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 82.1% |
| LMCA | — | 28.3% |
Math Qwen3 Max leads
MiniMax-M2.5: 25.8 (#242), Qwen3 Max: 37.6 (#94)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| LMArena Math | 1378 | 1447 |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| ProofBench | 4% | — |
| MATH Level 5 | — | 97.1% |
Knowledge Qwen3 Max leads
MiniMax-M2.5: 38.6 (#132), Qwen3 Max: 48.1 (#72)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| LMArena Expert | 1378 | 1457 |
| GPQA Diamond | — | 72.6% |
| SimpleQA Verified | — | 48.7% |
| Vectara Hallucination Rate | 9.1% | — |
Multilingual Qwen3 Max leads
MiniMax-M2.5: 47.3 (#147), Qwen3 Max: 53.9 (#58)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1338 | 1429 |
| LMArena Chinese | 1392 | 1479 |
| LMArena French | 1363 | 1450 |
| LMArena German | 1361 | 1463 |
| LMArena Japanese | 1169 | 1399 |
| LMArena Korean | 1232 | 1397 |
| LMArena Russian | 1359 | 1428 |
| LMArena Spanish | 1354 | 1463 |
Instruction Following Qwen3 Max leads
MiniMax-M2.5: 71.7 (#143), Qwen3 Max: 75.0 (#81)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1353 | 1420 |
Long Context Qwen3 Max leads
MiniMax-M2.5: 37.4 (#211), Qwen3 Max: 41.6 (#129)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| CL-bench | 11.4% | 14.5% |
| LMArena Longer Query | 1367 | 1439 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench Life | 6.3% | — |
Writing & Preference Qwen3 Max leads
MiniMax-M2.5: 54.1 (#149), Qwen3 Max: 62.6 (#72)
| Benchmark | MiniMax-M2.5 | Qwen3 Max |
|---|---|---|
| LMArena Text | 1359 | 1439 |
| LMArena Creative Writing | 1331 | 1401 |
| LMArena Multi-Turn | 1364 | 1446 |
| EQ-Bench Creative Writing | 1361 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Qwen3 Max?
Qwen3 Max is the stronger model overall, scoring 43.5 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 4.6× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2.5 or Qwen3 Max?
MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is MiniMax-M2.5 or Qwen3 Max better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Qwen3 Max does, with 262K tokens against 205K.
How many benchmarks do MiniMax-M2.5 and Qwen3 Max share?
23 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Qwen3 Max has 33.