Model comparison
Mistral vs Qwen1.5-72B
Mistral and Qwen1.5-72B score almost the same on the Noometry Index (29.9 vs 30.8), so choose on price, context window or the category you care about most.
Last verified . 17 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Mistral scores higher in 3 categories and Qwen1.5-72B in 5 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-72B leads 33.2 to 22.3.
- Qwen1.5-72B has downloadable open weights; the other is API-only.
Side by side
| Mistral | Qwen1.5-72B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 29.9 | 30.8 |
| Released | — | 2024-02-04 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral leads
Mistral: 33.8 (#250), Qwen1.5-72B: 31.9 (#277)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Coding | 1162 | 1165 |
| BigCodeBench Instruct | — | 33.2% |
| BigCodeBench Complete | — | 40.3% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 61.6% |
Reasoning Too close to call
Mistral: 22.2 (#200), Qwen1.5-72B: 22.2 (#203)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1148 |
Math Qwen1.5-72B leads
Mistral: 22.3 (#278), Qwen1.5-72B: 33.2 (#205)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1180 | 1164 |
| Omni-MATH | 7.2% | — |
Knowledge Mistral leads
Mistral: 16.6 (#288), Qwen1.5-72B: 11.5 (#300)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Expert | 1125 | 1136 |
| GPQA Diamond | — | 28.8% |
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
Multilingual Too close to call
Mistral: 32.8 (#254), Qwen1.5-72B: 33.2 (#253)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 1129 | 1135 |
| LMArena Chinese | 1109 | 1186 |
| LMArena French | 1180 | 1159 |
| LMArena German | 1155 | 1084 |
| LMArena Japanese | 1013 | 1061 |
| LMArena Korean | 1032 | 1050 |
| LMArena Russian | 1168 | 1104 |
| LMArena Spanish | 1143 | 1110 |
Instruction Following Qwen1.5-72B leads
Mistral: 52.6 (#288), Qwen1.5-72B: 59.3 (#256)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1152 | 1141 |
| IFEval | 56.8% | — |
Long Context Too close to call
Mistral: 35.0 (#245), Qwen1.5-72B: 35.1 (#243)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 1153 | 1157 |
Writing & Preference Too close to call
Mistral: 37.0 (#260), Qwen1.5-72B: 37.3 (#258)
| Benchmark | Mistral | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1165 | 1166 |
| LMArena Creative Writing | 1158 | 1137 |
| LMArena Multi-Turn | 1147 | 1160 |
| WildBench | 66% | — |
Frequently asked questions
Is Mistral better than Qwen1.5-72B?
Mistral and Qwen1.5-72B score almost the same on the Noometry Index (29.9 vs 30.8), so choose on price, context window or the category you care about most.
Is Mistral or Qwen1.5-72B better for coding?
Mistral scores higher on coding benchmarks: 33.8 versus 31.9 in the Noometry coding category.
How many benchmarks do Mistral and Qwen1.5-72B share?
17 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Qwen1.5-72B has 22.