Model comparison
DeepSeek-V3.1 vs Mistral
DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 29.4 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 17 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 8 categories and Mistral in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.0 to 16.3.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | Mistral | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 42.2 | 29.4 |
| Released | 2025-08-21 | — |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 39.8 (#140), Mistral: 33.2 (#249)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Coding | 1417 | 1162 |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.3 (#109), Mistral: 21.3 (#199)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1149 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.1 (#125), Mistral: 20.5 (#275)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Math | 1420 | 1180 |
| Omni-MATH | — | 7.2% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.0 (#86), Mistral: 16.3 (#283)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Expert | 1403 | 1125 |
| MMLU-Pro | — | 27.7% |
| Vectara Hallucination Rate | 5.5% | — |
| GPQA (HELM) | — | 30.3% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.8 (#100), Mistral: 33.0 (#249)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Non-English | 1400 | 1129 |
| LMArena Chinese | 1470 | 1109 |
| LMArena French | 1447 | 1179 |
| LMArena German | 1412 | 1155 |
| LMArena Japanese | 1374 | 1012 |
| LMArena Korean | 1337 | 1032 |
| LMArena Russian | 1405 | 1168 |
| LMArena Spanish | 1429 | 1143 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 74.1 (#105), Mistral: 52.9 (#283)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Instruction Following | 1400 | 1152 |
| IFEval | — | 56.8% |
Long Context DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.3 (#227), Mistral: 35.0 (#240)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Longer Query | 1422 | 1154 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.5 (#93), Mistral: 37.2 (#255)
| Benchmark | DeepSeek-V3.1 | Mistral |
|---|---|---|
| LMArena Text | 1420 | 1165 |
| LMArena Creative Writing | 1401 | 1158 |
| LMArena Multi-Turn | 1408 | 1147 |
| EQ-Bench Creative Writing | 1436 | — |
| WildBench | — | 66% |
Frequently asked questions
Is DeepSeek-V3.1 better than Mistral?
DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 29.4 on the Noometry Index.
Is DeepSeek-V3.1 or Mistral better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 39.8 versus 33.2 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Mistral share?
17 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Mistral has 22.