Model comparison
Mixtral 8x7B vs Qwen2-72B
Qwen2-72B is the stronger model overall, scoring 30.0 to 27.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Mixtral 8x7B scores higher in 1 category and Qwen2-72B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2-72B leads 30.2 to 18.8.
- The biggest single-benchmark swing is MATH Level 5: 10% for Mixtral 8x7B and 39.1% for Qwen2-72B.
Side by side
| Mixtral 8x7B | Qwen2-72B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 27.1 | 30.0 |
| Released | 2023-12-11 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 32K | — |
| Max output | 32K | — |
| Input $ / M tokens | $0.70 | — |
| Output $ / M tokens | $0.70 | — |
| Results tracked | 38 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mixtral 8x7B leads
Mixtral 8x7B: 32.8 (#269), Qwen2-72B: 29.1 (#310)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1126 | 1196 |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
| HumanEval+ | 39.6% | — |
| MBPP+ | 49.7% | — |
Agentic & Tool Use Not comparable
Mixtral 8x7B: —, Qwen2-72B: 17.0 (#146)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Mixtral 8x7B: 18.2 (#285), Qwen2-72B: 23.2 (#181)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1115 | 1191 |
| Epoch Capabilities Index | 118.47 | 125.28 |
| DTBench | 49.6% | — |
| Adversarial NLI | 55.2% | — |
| ForecastBench | 56.3 | — |
| HellaSwag | 86.7% | — |
| PIQA | 83.6% | — |
| WinoGrande | 77.2% | — |
Math Qwen2-72B leads
Mixtral 8x7B: 18.8 (#289), Qwen2-72B: 30.2 (#236)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Math | 1147 | 1235 |
| MATH Level 5 | 10% | 39.1% |
| Omni-MATH | 10.5% | — |
| GSM8K | 74.4% | — |
Knowledge Qwen2-72B leads
Mixtral 8x7B: 11.0 (#301), Qwen2-72B: 21.2 (#275)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 30.6% | 40.8% |
| LMArena Expert | 1088 | 1171 |
| MMLU | 70.6% | 82.4% |
| MMLU-Pro | 33.5% | — |
| GPQA (HELM) | 29.6% | — |
| ARC (AI2) Challenge | 87.3% | — |
| OpenBookQA | 85.8% | — |
| TriviaQA | 82.2% | — |
Multilingual Qwen2-72B leads
Mixtral 8x7B: 29.6 (#266), Qwen2-72B: 35.9 (#244)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1077 | 1176 |
| LMArena Chinese | 1055 | 1240 |
| LMArena French | 1166 | 1170 |
| LMArena German | 1114 | 1151 |
| LMArena Japanese | 931 | 1111 |
| LMArena Korean | 968 | 1083 |
| LMArena Russian | 1090 | 1169 |
| LMArena Spanish | 1111 | 1169 |
Instruction Following Qwen2-72B leads
Mixtral 8x7B: 51.0 (#297), Qwen2-72B: 61.7 (#241)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1109 | 1181 |
| IFEval | 57.5% | — |
Long Context Qwen2-72B leads
Mixtral 8x7B: 33.4 (#260), Qwen2-72B: 36.1 (#235)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1103 | 1192 |
Writing & Preference Qwen2-72B leads
Mixtral 8x7B: 34.2 (#270), Qwen2-72B: 40.8 (#241)
| Benchmark | Mixtral 8x7B | Qwen2-72B |
|---|---|---|
| LMArena Text | 1132 | 1203 |
| LMArena Creative Writing | 1109 | 1181 |
| LMArena Multi-Turn | 1115 | 1196 |
| WildBench | 67.3% | — |
Frequently asked questions
Is Mixtral 8x7B better than Qwen2-72B?
Qwen2-72B is the stronger model overall, scoring 30.0 to 27.1 on the Noometry Index.
Is Mixtral 8x7B or Qwen2-72B better for coding?
Mixtral 8x7B scores higher on coding benchmarks: 32.8 versus 29.1 in the Noometry coding category.
How many benchmarks do Mixtral 8x7B and Qwen2-72B share?
21 benchmarks have published results for both models. Mixtral 8x7B has 38 scored results on Noometry and Qwen2-72B has 26.