Model comparison
Mixtral 8x22B vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 27.1 on the Noometry Index.
Last verified . 28 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 28 benchmarks with published results for both. Mixtral 8x22B scores higher in 1 category and Qwen3 235B-A22B in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.6 to 15.1.
- The biggest single-benchmark swing is Omni-MATH: 16.3% for Mixtral 8x22B and 71.8% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Qwen3 235B-A22B accepts more context: 131K tokens versus 64K.
Side by side
| Mixtral 8x22B | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 27.1 | 43.5 |
| Released | 2024-04-17 | 2025-04 |
| Weights | Open | Open |
| Context window | 64K | 131K |
| Max output | 64K | 16K |
| Input $ / M tokens | $2 | $0.70 |
| Output $ / M tokens | $6 | $2.80 |
| Results tracked | 34 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Mixtral 8x22B: 24.2 (#329), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| WeirdML | 3.2% | 41% |
| LMArena Coding | 1166 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| BigCodeBench Instruct | 40.6% | — |
| BigCodeBench Complete | 50.2% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Mixtral 8x22B: 23.1 (#127), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Cybench | 7.5% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning Mixtral 8x22B leads
Mixtral 8x22B: 19.9 (#248), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1433 |
| DTBench | 55.1% | 80.3% |
| Epoch Capabilities Index | 122.03 | 143.85 |
| ForecastBench | 56.3 | 59.7 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 9% |
| LMCA | — | 29.3% |
Math Qwen3 235B-A22B leads
Mixtral 8x22B: 22.9 (#275), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| Omni-MATH | 16.3% | 71.8% |
| LMArena Math | 1184 | 1432 |
| MATH Level 5 | 24.2% | 68.9% |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Mixtral 8x22B: 15.1 (#293), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 34.1% | 80.1% |
| MMLU-Pro | 46% | 84.4% |
| GPQA (HELM) | 33.4% | 72.7% |
| LMArena Expert | 1113 | 1463 |
| SimpleQA Verified | — | 40.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| MMLU | 77.8% | — |
Multilingual Qwen3 235B-A22B leads
Mixtral 8x22B: 32.8 (#255), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1128 | 1409 |
| LMArena Chinese | 1116 | 1481 |
| LMArena French | 1166 | 1445 |
| LMArena German | 1141 | 1433 |
| LMArena Japanese | 1037 | 1399 |
| LMArena Korean | 1057 | 1391 |
| LMArena Russian | 1158 | 1411 |
| LMArena Spanish | 1151 | 1430 |
Instruction Following Qwen3 235B-A22B leads
Mixtral 8x22B: 57.7 (#266), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| IFEval | 72.4% | 83.5% |
| LMArena Instruction Following | 1147 | 1408 |
Long Context Qwen3 235B-A22B leads
Mixtral 8x22B: 34.7 (#247), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1144 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Mixtral 8x22B: 36.9 (#262), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Mixtral 8x22B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1162 | 1419 |
| LMArena Creative Writing | 1141 | 1384 |
| WildBench | 71.1% | 86.6% |
| LMArena Multi-Turn | 1130 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
Frequently asked questions
Is Mixtral 8x22B better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 27.1 on the Noometry Index.
Which is cheaper, Mixtral 8x22B 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; Mixtral 8x22B lists at $2 and $6.
Is Mixtral 8x22B or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
Qwen3 235B-A22B does, with 131K tokens against 64K.
How many benchmarks do Mixtral 8x22B and Qwen3 235B-A22B share?
28 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen3 235B-A22B has 49.