Model comparison
DeepSeek-V2.5 (Sep 2024) vs Mistral
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 29.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 17 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 7 categories and Mistral in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V2.5 (Sep 2024) leads 34.8 to 16.6.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Mistral | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 37.6 | 29.9 |
| Released | 2024-09-06 | — |
| Weights | Open | Proprietary |
| 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
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Mistral: 33.8 (#250)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Coding | 1309 | 1162 |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Mistral: 22.2 (#200)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1149 |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Mistral: 22.3 (#278)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Math | 1288 | 1180 |
| Omni-MATH | — | 7.2% |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Mistral: 16.6 (#288)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Expert | 1266 | 1125 |
| MMLU-Pro | — | 27.7% |
| GPQA (HELM) | — | 30.3% |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Mistral: 32.8 (#254)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Non-English | 1273 | 1129 |
| LMArena Chinese | 1318 | 1109 |
| LMArena French | 1289 | 1180 |
| LMArena German | 1258 | 1155 |
| LMArena Japanese | 1228 | 1013 |
| LMArena Korean | 1209 | 1032 |
| LMArena Russian | 1289 | 1168 |
| LMArena Spanish | 1248 | 1143 |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Mistral: 52.6 (#288)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Instruction Following | 1280 | 1152 |
| IFEval | — | 56.8% |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Mistral: 35.0 (#245)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Longer Query | 1301 | 1153 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Mistral: 37.0 (#260)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Mistral |
|---|---|---|
| LMArena Text | 1294 | 1165 |
| LMArena Creative Writing | 1285 | 1158 |
| LMArena Multi-Turn | 1297 | 1147 |
| WildBench | — | 66% |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Mistral?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 29.9 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Mistral better for coding?
Mistral scores higher on coding benchmarks: 33.8 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Mistral share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Mistral has 22.