Model comparison
Mistral vs Qwen-14B
Qwen-14B is the stronger model overall, scoring 31.4 to 29.9 on the Noometry Index.
Last verified . 10 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. Mistral scores higher in 6 categories and Qwen-14B in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral leads 37.0 to 27.6.
- Qwen-14B has downloadable open weights; the other is API-only.
Side by side
| Mistral | Qwen-14B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 29.9 | 31.4 |
| Released | — | 2023-09-24 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral leads
Mistral: 33.8 (#250), Qwen-14B: 31.2 (#288)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Coding | 1162 | 1071 |
Reasoning Mistral leads
Mistral: 22.2 (#200), Qwen-14B: 19.6 (#257)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1027 |
| BIG-Bench Hard | — | 55% |
| Epoch Capabilities Index | — | 113.03 |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
Math Qwen-14B leads
Mistral: 22.3 (#278), Qwen-14B: 31.2 (#227)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Math | 1180 | 1068 |
| Omni-MATH | 7.2% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
Mistral: 16.6 (#288), Qwen-14B: —
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
| LMArena Expert | 1125 | — |
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
| MMLU | — | 66.3% |
Multilingual Mistral leads
Mistral: 32.8 (#254), Qwen-14B: 27.5 (#275)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1129 | 1041 |
| LMArena Chinese | 1109 | 1077 |
| LMArena French | 1180 | — |
| LMArena German | 1155 | — |
| LMArena Japanese | 1013 | — |
| LMArena Korean | 1032 | — |
| LMArena Russian | 1168 | — |
| LMArena Spanish | 1143 | — |
Instruction Following Too close to call
Mistral: 52.6 (#288), Qwen-14B: 52.4 (#289)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1152 | 1031 |
| IFEval | 56.8% | — |
Long Context Mistral leads
Mistral: 35.0 (#245), Qwen-14B: 31.3 (#280)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1153 | 1028 |
Writing & Preference Mistral leads
Mistral: 37.0 (#260), Qwen-14B: 27.6 (#299)
| Benchmark | Mistral | Qwen-14B |
|---|---|---|
| LMArena Text | 1165 | 1051 |
| LMArena Creative Writing | 1158 | 1028 |
| LMArena Multi-Turn | 1147 | 1022 |
| WildBench | 66% | — |
Frequently asked questions
Is Mistral better than Qwen-14B?
Qwen-14B is the stronger model overall, scoring 31.4 to 29.9 on the Noometry Index.
Is Mistral or Qwen-14B better for coding?
Mistral scores higher on coding benchmarks: 33.8 versus 31.2 in the Noometry coding category.
How many benchmarks do Mistral and Qwen-14B share?
10 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Qwen-14B has 18.