Model comparison
Mixtral 8x7B vs Qwen1.5-72B
Qwen1.5-72B is the stronger model overall, scoring 30.8 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 8x7B scores higher in 1 category and Qwen1.5-72B in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-72B leads 33.2 to 18.8.
Side by side
| Mixtral 8x7B | Qwen1.5-72B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 27.1 | 30.8 |
| Released | 2023-12-11 | 2024-02-04 |
| Weights | Open | Open |
| Context window | 32K | — |
| Max output | 32K | — |
| Input $ / M tokens | $0.70 | — |
| Output $ / M tokens | $0.70 | — |
| Results tracked | 38 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mixtral 8x7B: 32.8 (#269), Qwen1.5-72B: 31.9 (#277)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Coding | 1126 | 1165 |
| HumanEval+ | 39.6% | 59.1% |
| MBPP+ | 49.7% | 61.6% |
| BigCodeBench Instruct | — | 33.2% |
| BigCodeBench Complete | — | 40.3% |
Reasoning Qwen1.5-72B leads
Mixtral 8x7B: 18.2 (#285), Qwen1.5-72B: 22.2 (#203)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1115 | 1148 |
| DTBench | 49.6% | — |
| Adversarial NLI | 55.2% | — |
| Epoch Capabilities Index | 118.47 | — |
| ForecastBench | 56.3 | — |
| HellaSwag | 86.7% | — |
| PIQA | 83.6% | — |
| WinoGrande | 77.2% | — |
Math Qwen1.5-72B leads
Mixtral 8x7B: 18.8 (#289), Qwen1.5-72B: 33.2 (#205)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1147 | 1164 |
| Omni-MATH | 10.5% | — |
| MATH Level 5 | 10% | — |
| GSM8K | 74.4% | — |
Knowledge Too close to call
Mixtral 8x7B: 11.0 (#301), Qwen1.5-72B: 11.5 (#300)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| GPQA Diamond | 30.6% | 28.8% |
| LMArena Expert | 1088 | 1136 |
| MMLU-Pro | 33.5% | — |
| GPQA (HELM) | 29.6% | — |
| ARC (AI2) Challenge | 87.3% | — |
| MMLU | 70.6% | — |
| OpenBookQA | 85.8% | — |
| TriviaQA | 82.2% | — |
Multilingual Qwen1.5-72B leads
Mixtral 8x7B: 29.6 (#266), Qwen1.5-72B: 33.2 (#253)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 1077 | 1135 |
| LMArena Chinese | 1055 | 1186 |
| LMArena French | 1166 | 1159 |
| LMArena German | 1114 | 1084 |
| LMArena Japanese | 931 | 1061 |
| LMArena Korean | 968 | 1050 |
| LMArena Russian | 1090 | 1104 |
| LMArena Spanish | 1111 | 1110 |
Instruction Following Qwen1.5-72B leads
Mixtral 8x7B: 51.0 (#297), Qwen1.5-72B: 59.3 (#256)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1109 | 1141 |
| IFEval | 57.5% | — |
Long Context Qwen1.5-72B leads
Mixtral 8x7B: 33.4 (#260), Qwen1.5-72B: 35.1 (#243)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 1103 | 1157 |
Writing & Preference Qwen1.5-72B leads
Mixtral 8x7B: 34.2 (#270), Qwen1.5-72B: 37.3 (#258)
| Benchmark | Mixtral 8x7B | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1132 | 1166 |
| LMArena Creative Writing | 1109 | 1137 |
| LMArena Multi-Turn | 1115 | 1160 |
| WildBench | 67.3% | — |
Frequently asked questions
Is Mixtral 8x7B better than Qwen1.5-72B?
Qwen1.5-72B is the stronger model overall, scoring 30.8 to 27.1 on the Noometry Index.
Is Mixtral 8x7B or Qwen1.5-72B better for coding?
They score almost the same on coding (32.8 vs 31.9); test both on your own repository before choosing.
How many benchmarks do Mixtral 8x7B and Qwen1.5-72B share?
20 benchmarks have published results for both models. Mixtral 8x7B has 38 scored results on Noometry and Qwen1.5-72B has 22.