Model comparison
Mixtral 8x22B vs Qwen2.5 7B Instruct
Qwen2.5 7B Instruct is the stronger model overall, scoring 29.0 to 27.1 on the Noometry Index.
Last verified . 11 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. Mixtral 8x22B scores higher in 2 categories and Qwen2.5 7B Instruct in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen2.5 7B Instruct leads 36.5 to 24.2.
- The biggest single-benchmark swing is Omni-MATH: 16.3% for Mixtral 8x22B and 29.4% for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Qwen2.5 7B Instruct accepts more context: 131K tokens versus 64K.
Side by side
| Mixtral 8x22B | Qwen2.5 7B Instruct | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 27.1 | 29.0 |
| Released | 2024-04-17 | 2024-09 |
| Weights | Open | Open |
| Context window | 64K | 131K |
| Max output | 64K | 8K |
| Input $ / M tokens | $2 | $0.17 |
| Output $ / M tokens | $6 | $0.70 |
| Results tracked | 34 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Mixtral 8x22B: 24.2 (#329), Qwen2.5 7B Instruct: 36.5 (#208)
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| BigCodeBench Instruct | 40.6% | 37.6% |
| BigCodeBench Complete | 50.2% | 46.1% |
| WeirdML | 3.2% | — |
| LMArena Coding | 1166 | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Too close to call
Mixtral 8x22B: 23.1 (#127), Qwen2.5 7B Instruct: 23.8 (#124)
Reasoning Mixtral 8x22B leads
Mixtral 8x22B: 19.9 (#248), Qwen2.5 7B Instruct: 14.8 (#322)
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| DTBench | 55.1% | 47.7% |
| Epoch Capabilities Index | 122.03 | 118.51 |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | 1150 | — |
| LMCA | — | 6.4% |
| ForecastBench | 56.3 | — |
Math Mixtral 8x22B leads
Mixtral 8x22B: 22.9 (#275), Qwen2.5 7B Instruct: 12.6 (#306)
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| Omni-MATH | 16.3% | 29.4% |
| OTIS Mock AIME 2024-2025 | — | 2.5% |
| LMArena Math | 1184 | — |
| MATH Level 5 | 24.2% | — |
Knowledge Qwen2.5 7B Instruct leads
Mixtral 8x22B: 15.1 (#293), Qwen2.5 7B Instruct: 17.0 (#286)
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 34.1% | 35.5% |
| MMLU-Pro | 46% | 53.9% |
| GPQA (HELM) | 33.4% | 34.1% |
| MMLU | 77.8% | 72.9% |
| LMArena Expert | 1113 | — |
Multilingual Not comparable
Mixtral 8x22B: 32.8 (#255), Qwen2.5 7B Instruct: —
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1128 | — |
| LMArena Chinese | 1116 | — |
| LMArena French | 1166 | — |
| LMArena German | 1141 | — |
| LMArena Japanese | 1037 | — |
| LMArena Korean | 1057 | — |
| LMArena Russian | 1158 | — |
| LMArena Spanish | 1151 | — |
Instruction Following Qwen2.5 7B Instruct leads
Mixtral 8x22B: 57.7 (#266), Qwen2.5 7B Instruct: 63.2 (#231)
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | 72.4% | 74.1% |
| LMArena Instruction Following | 1147 | — |
Long Context Not comparable
Mixtral 8x22B: 34.7 (#247), Qwen2.5 7B Instruct: —
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1144 | — |
Writing & Preference Qwen2.5 7B Instruct leads
Mixtral 8x22B: 36.9 (#262), Qwen2.5 7B Instruct: 48.8 (#195)
| Benchmark | Mixtral 8x22B | Qwen2.5 7B Instruct |
|---|---|---|
| WildBench | 71.1% | 73.1% |
| LMArena Text | 1162 | — |
| LMArena Creative Writing | 1141 | — |
| LMArena Multi-Turn | 1130 | — |
Frequently asked questions
Is Mixtral 8x22B better than Qwen2.5 7B Instruct?
Qwen2.5 7B Instruct is the stronger model overall, scoring 29.0 to 27.1 on the Noometry Index.
Which is cheaper, Mixtral 8x22B or Qwen2.5 7B Instruct?
Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Mixtral 8x22B or Qwen2.5 7B Instruct better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 36.5 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5 7B Instruct does, with 131K tokens against 64K.
How many benchmarks do Mixtral 8x22B and Qwen2.5 7B Instruct share?
11 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen2.5 7B Instruct has 15.