Model comparison
Mixtral 8x22B vs Qwen3-30B-A3B
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 26.7 on the Noometry Index.
Last verified . 21 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Mixtral 8x22B scores higher in 1 category and Qwen3-30B-A3B in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3-30B-A3B leads 41.7 to 15.1.
- The biggest single-benchmark swing is GPQA Diamond: 34.1% for Mixtral 8x22B and 70.1% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Mixtral 8x22B accepts more context: 64K tokens versus 41K.
Side by side
| Mixtral 8x22B | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 26.7 | 38.5 |
| Released | 2024-04-17 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 64K | 41K |
| Max output | 64K | 16K |
| Input $ / M tokens | $2 | $0.12 |
| Output $ / M tokens | $6 | $0.50 |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3-30B-A3B leads
Mixtral 8x22B: 23.8 (#324), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| WeirdML | 3.2% | 29.8% |
| LMArena Coding | 1166 | 1416 |
| SciCode | — | 33.3% |
| BigCodeBench Instruct | 40.6% | — |
| BigCodeBench Complete | 50.2% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Qwen3-30B-A3B leads
Mixtral 8x22B: 22.8 (#119), Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
| Cybench | 7.5% | — |
Reasoning Qwen3-30B-A3B leads
Mixtral 8x22B: 19.1 (#246), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1398 |
| DTBench | 55.1% | 69.3% |
| Epoch Capabilities Index | 122 | 139.63 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 8% |
| LMCA | — | 22.4% |
| ForecastBench | 56.3 | — |
Math Qwen3-30B-A3B leads
Mixtral 8x22B: 21.2 (#271), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Math | 1184 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| OTIS Mock AIME 2024-2025 | — | 70.3% |
| Omni-MATH | 16.3% | — |
| MATH Level 5 | 24.2% | — |
Knowledge Qwen3-30B-A3B leads
Mixtral 8x22B: 15.1 (#288), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 34.1% | 70.1% |
| LMArena Expert | 1113 | 1395 |
| MMLU-Pro | 46% | — |
| Confabulations | — | 12.3% |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |
Multilingual Qwen3-30B-A3B leads
Mixtral 8x22B: 32.9 (#250), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1128 | 1372 |
| LMArena Chinese | 1116 | 1434 |
| LMArena French | 1165 | 1419 |
| LMArena German | 1141 | 1379 |
| LMArena Japanese | 1037 | 1338 |
| LMArena Korean | 1056 | 1333 |
| LMArena Russian | 1158 | 1369 |
| LMArena Spanish | 1150 | 1403 |
Instruction Following Qwen3-30B-A3B leads
Mixtral 8x22B: 57.9 (#261), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1147 | 1364 |
| IFEval | 72.4% | — |
Long Context Mixtral 8x22B leads
Mixtral 8x22B: 34.6 (#242), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1144 | 1379 |
| Fiction.LiveBench | — | 40.6% |
Writing & Preference Qwen3-30B-A3B leads
Mixtral 8x22B: 37.1 (#257), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | Mixtral 8x22B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1162 | 1384 |
| LMArena Creative Writing | 1141 | 1318 |
| LMArena Multi-Turn | 1130 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
| WildBench | 71.1% | — |
Frequently asked questions
Is Mixtral 8x22B better than Qwen3-30B-A3B?
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 26.7 on the Noometry Index.
Which is cheaper, Mixtral 8x22B or Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Mixtral 8x22B or Qwen3-30B-A3B better for coding?
Qwen3-30B-A3B scores higher on coding benchmarks: 37.0 versus 23.8 in the Noometry coding category.
Which has the bigger context window?
Mixtral 8x22B does, with 64K tokens against 41K.
How many benchmarks do Mixtral 8x22B and Qwen3-30B-A3B share?
21 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen3-30B-A3B has 32.