Model comparison
Mistral vs Qwen1.5-14B
Qwen1.5-14B is the stronger model overall, scoring 32.1 to 29.4 on the Noometry Index.
Last verified . 16 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 16 benchmarks with published results for both. Mistral scores higher in 5 categories and Qwen1.5-14B in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen1.5-14B leads 29.3 to 16.3.
- Qwen1.5-14B has downloadable open weights; the other is API-only.
Side by side
| Mistral | Qwen1.5-14B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 29.4 | 32.1 |
| Released | — | 2024-02-04 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral: 33.2 (#249), Qwen1.5-14B: 32.5 (#259)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Coding | 1162 | 1138 |
Reasoning Too close to call
Mistral: 21.3 (#199), Qwen1.5-14B: 20.5 (#217)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1113 |
Math Qwen1.5-14B leads
Mistral: 20.5 (#275), Qwen1.5-14B: 29.9 (#214)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Math | 1180 | 1125 |
| Omni-MATH | 7.2% | — |
Knowledge Qwen1.5-14B leads
Mistral: 16.3 (#283), Qwen1.5-14B: 29.3 (#230)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Expert | 1125 | 1094 |
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
| MMLU | — | 68.6% |
Multilingual Mistral leads
Mistral: 33.0 (#249), Qwen1.5-14B: 30.8 (#257)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Non-English | 1129 | 1095 |
| LMArena Chinese | 1109 | 1147 |
| LMArena French | 1179 | 1115 |
| LMArena German | 1155 | 1043 |
| LMArena Japanese | 1012 | 1019 |
| LMArena Russian | 1168 | 1046 |
| LMArena Spanish | 1143 | 1085 |
| LMArena Korean | 1032 | — |
Instruction Following Qwen1.5-14B leads
Mistral: 52.9 (#283), Qwen1.5-14B: 57.1 (#266)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Instruction Following | 1152 | 1102 |
| IFEval | 56.8% | — |
Long Context Mistral leads
Mistral: 35.0 (#240), Qwen1.5-14B: 33.7 (#252)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Longer Query | 1154 | 1113 |
Writing & Preference Mistral leads
Mistral: 37.2 (#255), Qwen1.5-14B: 33.7 (#270)
| Benchmark | Mistral | Qwen1.5-14B |
|---|---|---|
| LMArena Text | 1165 | 1128 |
| LMArena Creative Writing | 1158 | 1091 |
| LMArena Multi-Turn | 1147 | 1110 |
| WildBench | 66% | — |
Frequently asked questions
Is Mistral better than Qwen1.5-14B?
Qwen1.5-14B is the stronger model overall, scoring 32.1 to 29.4 on the Noometry Index.
Is Mistral or Qwen1.5-14B better for coding?
They score almost the same on coding (33.2 vs 32.5); test both on your own repository before choosing.
How many benchmarks do Mistral and Qwen1.5-14B share?
16 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Qwen1.5-14B has 17.