Model comparison
Mistral Small vs Qwen3 Max
Qwen3 Max is the stronger model overall, scoring 43.7 to 33.4 on the Noometry Index. Mistral Small costs 9.1× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Mistral Small scores higher in 0 categories and Qwen3 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 Max leads 38.7 to 16.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 5.8% for Mistral Small and 73.3% for Qwen3 Max.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| Mistral Small | Qwen3 Max | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 33.4 | 43.7 |
| Released | 2024-02-26 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 262K | 262K |
| Max output | 256K | 66K |
| Input $ / M tokens | $0.15 | $1.20 |
| Output $ / M tokens | $0.60 | $6 |
| Results tracked | 39 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 Max leads
Mistral Small: 34.0 (#247), Qwen3 Max: 43.0 (#93)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1362 | 1456 |
| ALE-Bench | 497.62 | 370.45 |
| SciCode | 26.5% | — |
| BigCodeBench Instruct | 36.1% | — |
| LiveBench Coding | 36.2% | — |
| BigCodeBench Complete | 46.6% | — |
Agentic & Tool Use Not comparable
Mistral Small: 28.1 (#93), Qwen3 Max: —
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.1% | — |
| Vending-Bench 2 | — | 71.56 |
Reasoning Qwen3 Max leads
Mistral Small: 19.8 (#250), Qwen3 Max: 22.6 (#190)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 37.8% | 72.5% |
| LMArena Hard Prompts | 1335 | 1448 |
| DTBench | 70.9% | 82.1% |
| LMCA | 20.6% | 28.3% |
| NYT Connections (extended) | — | 30.1% |
| CritPt | 0% | — |
| Chess Puzzles | — | 4% |
| LiveBench Reasoning | 44.8% | — |
| Mystery Game Puzzles | — | 5% |
| LiveBench Data Analysis | 53.7% | — |
| Epoch Capabilities Index | — | 142.38 |
| LiveBench | 44% | — |
Math Qwen3 Max leads
Mistral Small: 16.4 (#293), Qwen3 Max: 38.7 (#131)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.8% | 73.3% |
| LMArena Math | 1341 | 1446 |
| MATH Level 5 | 46.8% | 97.1% |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| LiveBench Math | 39.9% | — |
Knowledge Qwen3 Max leads
Mistral Small: 31.0 (#222), Qwen3 Max: 48.1 (#78)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 47.5% | 72.6% |
| LMArena Expert | 1291 | 1455 |
| SimpleQA Verified | — | 48.7% |
| Vectara Hallucination Rate | 5.1% | — |
| MMLU | 68.7% | — |
Multimodal Not comparable
Mistral Small: 33.5 (#96), Qwen3 Max: —
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| LMArena Vision | 1142 | — |
Multilingual Qwen3 Max leads
Mistral Small: 45.5 (#169), Qwen3 Max: 53.7 (#62)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1315 | 1429 |
| LMArena Chinese | 1340 | 1478 |
| LMArena French | 1337 | 1449 |
| LMArena German | 1340 | 1463 |
| LMArena Japanese | 1275 | 1397 |
| LMArena Korean | 1259 | 1399 |
| LMArena Russian | 1324 | 1428 |
| LMArena Spanish | 1346 | 1462 |
Instruction Following Qwen3 Max leads
Mistral Small: 66.4 (#209), Qwen3 Max: 74.8 (#87)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1310 | 1419 |
| LiveBench Instruction Following | 63.7% | — |
Long Context Qwen3 Max leads
Mistral Small: 40.4 (#156), Qwen3 Max: 41.6 (#134)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1327 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |
Writing & Preference Qwen3 Max leads
Mistral Small: 52.5 (#171), Qwen3 Max: 62.4 (#76)
| Benchmark | Mistral Small | Qwen3 Max |
|---|---|---|
| LMArena Text | 1338 | 1439 |
| LMArena Creative Writing | 1305 | 1402 |
| LMArena Multi-Turn | 1344 | 1446 |
| LiveBench Language | 30.5% | — |
Frequently asked questions
Is Mistral Small better than Qwen3 Max?
Qwen3 Max is the stronger model overall, scoring 43.7 to 33.4 on the Noometry Index. Mistral Small costs 9.1× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Which is cheaper, Mistral Small or Qwen3 Max?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is Mistral Small or Qwen3 Max better for coding?
Qwen3 Max scores higher on coding benchmarks: 43.0 versus 34.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Mistral Small and Qwen3 Max share?
24 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Qwen3 Max has 33.