Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Mistral
Mistral has enough public results to be ranked (#297); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
DeepSeek
Mistral AISummary
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 39.8 to 33.2.
- DeepSeek-V2 (MoE-236B, May 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Mistral | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 39.7 | 29.4 |
| Released | 2024-05-07 | — |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2 (MoE-236B, May 2024) leads
DeepSeek-V2 (MoE-236B, May 2024): 39.8 (#141), Mistral: 33.2 (#249)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| BigCodeBench Instruct | 48.9% | — |
| LMArena Coding | — | 1162 |
| BigCodeBench Complete | 59.4% | — |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 21.3 (#199)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| LMArena Hard Prompts | — | 1149 |
| BIG-Bench Hard | 78.8% | — |
| Epoch Capabilities Index | 124.78 | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 20.5 (#275)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| Omni-MATH | — | 7.2% |
| LMArena Math | — | 1180 |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 16.3 (#283)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| MMLU-Pro | — | 27.7% |
| GPQA (HELM) | — | 30.3% |
| LMArena Expert | — | 1125 |
| ARC (AI2) Challenge | 92.2% | — |
| MMLU | 78.4% | — |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 33.0 (#249)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| LMArena Non-English | — | 1129 |
| LMArena Chinese | — | 1109 |
| LMArena French | — | 1179 |
| LMArena German | — | 1155 |
| LMArena Japanese | — | 1012 |
| LMArena Korean | — | 1032 |
| LMArena Russian | — | 1168 |
| LMArena Spanish | — | 1143 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 52.9 (#283)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| IFEval | — | 56.8% |
| LMArena Instruction Following | — | 1152 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 35.0 (#240)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| LMArena Longer Query | — | 1154 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mistral: 37.2 (#255)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mistral |
|---|---|---|
| LMArena Text | — | 1165 |
| LMArena Creative Writing | — | 1158 |
| WildBench | — | 66% |
| LMArena Multi-Turn | — | 1147 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Mistral?
Mistral has enough public results to be ranked (#297); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Mistral better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 39.8 versus 33.2 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Mistral share?
0 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Mistral has 22.