Model comparison
Mistral Large vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 31.9 on the Noometry Index.
Last verified . 38 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 38 benchmarks with published results for both. Mistral Large scores higher in 1 category and Qwen3 235B-A22B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 235B-A22B leads 50.4 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Mistral Large and 86.7% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $2 / $6 for Mistral Large.
Side by side
| Mistral Large | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 31.9 | 43.5 |
| Released | 2024-02-26 | 2025-04 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 16K | 16K |
| Input $ / M tokens | $2 | $0.70 |
| Output $ / M tokens | $6 | $2.80 |
| Results tracked | 51 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Mistral Large: 34.3 (#240), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 36.2% | 42.4% |
| LMArena Coding | 1277 | 1445 |
| Aider Polyglot | — | 59.6% |
| WeirdML | — | 41% |
| BigCodeBench Instruct | 30% | — |
| LiveBench Coding | 47.1% | — |
| BigCodeBench Complete | 38.3% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Mistral Large: 28.6 (#89), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning Too close to call
Mistral Large: 15.8 (#310), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| SimpleBench | 22.5% | 31% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1257 | 1433 |
| DTBench | 65.1% | 80.3% |
| LMCA | 16.7% | 29.3% |
| Epoch Capabilities Index | 128.52 | 143.85 |
| ForecastBench | 57.1 | 59.7 |
| ARC-AGI-2 | — | 1.3% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| Chess Puzzles | — | 12% |
| LiveBench Reasoning | 43.5% | — |
| Mystery Game Puzzles | — | 9% |
| LiveBench Data Analysis | 50.1% | — |
| LiveBench | 48.4% | — |
Math Qwen3 235B-A22B leads
Mistral Large: 18.2 (#291), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 86.7% |
| Omni-MATH | 28.1% | 71.8% |
| LMArena Math | 1262 | 1432 |
| MATH Level 5 | 50.3% | 68.9% |
| FrontierMath (Feb 2025 set) | 0.3% | 8.5% |
| LiveBench Math | 42.5% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Mistral Large: 30.1 (#230), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 51.3% | 80.1% |
| MMLU-Pro | 59.9% | 84.4% |
| Confabulations | 21.4% | 15.6% |
| Vectara Hallucination Rate | 4.5% | 9.3% |
| GPQA (HELM) | 43.5% | 72.7% |
| LMArena Expert | 1232 | 1463 |
| SimpleQA Verified | — | 40.4% |
| MMLU | 80% | — |
Multilingual Qwen3 235B-A22B leads
Mistral Large: 40.0 (#219), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1237 | 1409 |
| LMArena Chinese | 1240 | 1481 |
| LMArena French | 1325 | 1445 |
| LMArena German | 1254 | 1433 |
| LMArena Japanese | 1188 | 1399 |
| LMArena Korean | 1202 | 1391 |
| LMArena Russian | 1257 | 1411 |
| LMArena Spanish | 1268 | 1430 |
Instruction Following Qwen3 235B-A22B leads
Mistral Large: 67.9 (#191), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| IFEval | 87.7% | 83.5% |
| LMArena Instruction Following | 1249 | 1408 |
| LiveBench Instruction Following | 67.9% | — |
Long Context Qwen3 235B-A22B leads
Mistral Large: 38.3 (#199), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1261 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Mistral Large: 40.7 (#242), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Mistral Large | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1266 | 1419 |
| LMArena Creative Writing | 1243 | 1384 |
| Short-Story Creative Writing | 69% | 83% |
| EQ-Bench Creative Writing | 985 | 1366 |
| WildBench | 80.1% | 86.6% |
| LMArena Multi-Turn | 1260 | 1432 |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 31.9 on the Noometry Index.
Which is cheaper, Mistral Large or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Mistral Large lists at $2 and $6.
Is Mistral Large or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do Mistral Large and Qwen3 235B-A22B share?
38 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Qwen3 235B-A22B has 49.