Model comparison
Mixtral 8x22B vs Qwen3.5-Flash
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 27.1 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Mixtral 8x22B scores higher in 0 categories and Qwen3.5-Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5-Flash leads 43.2 to 15.1.
- The biggest single-benchmark swing is GPQA Diamond: 34.1% for Mixtral 8x22B and 82.3% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Qwen3.5-Flash accepts more context: 1M tokens versus 64K.
- Mixtral 8x22B has downloadable open weights; the other is API-only.
Side by side
| Mixtral 8x22B | Qwen3.5-Flash | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 27.1 | 42.5 |
| Released | 2024-04-17 | 2026-02-23 |
| Weights | Open | Proprietary |
| Context window | 64K | 1M |
| Max output | 64K | 66K |
| Input $ / M tokens | $2 | $0.10 |
| Output $ / M tokens | $6 | $0.40 |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5-Flash leads
Mixtral 8x22B: 24.2 (#329), Qwen3.5-Flash: 34.2 (#242)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Coding | 1166 | 1412 |
| LMArena WebDev | — | 1244 |
| WeirdML | 3.2% | — |
| BigCodeBench Instruct | 40.6% | — |
| BigCodeBench Complete | 50.2% | — |
| ALE-Bench | — | 221.8 |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Not comparable
Mixtral 8x22B: 23.1 (#127), Qwen3.5-Flash: —
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| Cybench | 7.5% | — |
| Vending-Bench 2 | — | 462.69 |
Reasoning Qwen3.5-Flash leads
Mixtral 8x22B: 19.9 (#248), Qwen3.5-Flash: 33.7 (#72)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1403 |
| DTBench | 55.1% | 82.9% |
| Epoch Capabilities Index | 122.03 | 143.98 |
| Chess Puzzles | — | 21% |
| Mystery Game Puzzles | — | 20% |
| LMCA | — | 29.1% |
| ForecastBench | 56.3 | — |
Math Qwen3.5-Flash leads
Mixtral 8x22B: 22.9 (#275), Qwen3.5-Flash: 37.4 (#158)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Math | 1184 | 1407 |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| OTIS Mock AIME 2024-2025 | — | 84.4% |
| Omni-MATH | 16.3% | — |
| MATH Level 5 | 24.2% | — |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3.5-Flash leads
Mixtral 8x22B: 15.1 (#293), Qwen3.5-Flash: 43.2 (#93)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 34.1% | 82.3% |
| LMArena Expert | 1113 | 1407 |
| SimpleQA Verified | — | 20.3% |
| MMLU-Pro | 46% | — |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |
Multilingual Qwen3.5-Flash leads
Mixtral 8x22B: 32.8 (#255), Qwen3.5-Flash: 50.5 (#121)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1128 | 1385 |
| LMArena Chinese | 1116 | 1446 |
| LMArena French | 1166 | 1412 |
| LMArena German | 1141 | 1390 |
| LMArena Japanese | 1037 | 1368 |
| LMArena Korean | 1057 | 1344 |
| LMArena Russian | 1158 | 1379 |
| LMArena Spanish | 1151 | 1400 |
Instruction Following Qwen3.5-Flash leads
Mixtral 8x22B: 57.7 (#266), Qwen3.5-Flash: 72.6 (#139)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1147 | 1374 |
| IFEval | 72.4% | — |
Long Context Qwen3.5-Flash leads
Mixtral 8x22B: 34.7 (#247), Qwen3.5-Flash: 42.4 (#124)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1144 | 1392 |
Writing & Preference Qwen3.5-Flash leads
Mixtral 8x22B: 36.9 (#262), Qwen3.5-Flash: 57.9 (#122)
| Benchmark | Mixtral 8x22B | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1162 | 1397 |
| LMArena Creative Writing | 1141 | 1343 |
| LMArena Multi-Turn | 1130 | 1393 |
| WildBench | 71.1% | — |
Frequently asked questions
Is Mixtral 8x22B better than Qwen3.5-Flash?
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 27.1 on the Noometry Index.
Which is cheaper, Mixtral 8x22B or Qwen3.5-Flash?
Qwen3.5-Flash is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Mixtral 8x22B or Qwen3.5-Flash better for coding?
Qwen3.5-Flash scores higher on coding benchmarks: 34.2 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5-Flash does, with 1M tokens against 64K.
How many benchmarks do Mixtral 8x22B and Qwen3.5-Flash share?
20 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen3.5-Flash has 32.