Model comparison
DeepSeek LLM 67B vs Mistral
Mistral is the stronger model overall, scoring 29.4 to 24.8 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 10 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 1 category and Mistral in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Mistral leads 20.5 to 8.1.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek LLM 67B | Mistral | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 24.8 | 29.4 |
| Released | 2023-11-29 | — |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral leads
DeepSeek LLM 67B: 31.3 (#273), Mistral: 33.2 (#249)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Coding | 1096 | 1162 |
Reasoning Mistral leads
DeepSeek LLM 67B: 15.9 (#300), Mistral: 21.3 (#199)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1149 |
| Chess Puzzles | 0% | — |
| Epoch Capabilities Index | 110.54 | — |
Math Mistral leads
DeepSeek LLM 67B: 8.1 (#318), Mistral: 20.5 (#275)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Math | 1108 | 1180 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| Omni-MATH | — | 7.2% |
| MATH Level 5 | 6.4% | — |
Knowledge Mistral leads
DeepSeek LLM 67B: 7.1 (#308), Mistral: 16.3 (#283)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| MMLU-Pro | — | 27.7% |
| GPQA (HELM) | — | 30.3% |
| LMArena Expert | — | 1125 |
Multilingual Mistral leads
DeepSeek LLM 67B: 29.5 (#262), Mistral: 33.0 (#249)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Non-English | 1073 | 1129 |
| LMArena Chinese | 1132 | 1109 |
| LMArena French | — | 1179 |
| LMArena German | — | 1155 |
| LMArena Japanese | — | 1012 |
| LMArena Korean | — | 1032 |
| LMArena Russian | — | 1168 |
| LMArena Spanish | — | 1143 |
Instruction Following DeepSeek LLM 67B leads
DeepSeek LLM 67B: 55.7 (#272), Mistral: 52.9 (#283)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Instruction Following | 1079 | 1152 |
| IFEval | — | 56.8% |
Long Context Mistral leads
DeepSeek LLM 67B: 33.1 (#260), Mistral: 35.0 (#240)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Longer Query | 1092 | 1154 |
Writing & Preference Mistral leads
DeepSeek LLM 67B: 31.8 (#277), Mistral: 37.2 (#255)
| Benchmark | DeepSeek LLM 67B | Mistral |
|---|---|---|
| LMArena Text | 1105 | 1165 |
| LMArena Creative Writing | 1067 | 1158 |
| LMArena Multi-Turn | 1082 | 1147 |
| WildBench | — | 66% |
Frequently asked questions
Is DeepSeek LLM 67B better than Mistral?
Mistral is the stronger model overall, scoring 29.4 to 24.8 on the Noometry Index.
Is DeepSeek LLM 67B or Mistral better for coding?
Mistral scores higher on coding benchmarks: 33.2 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Mistral share?
10 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Mistral has 22.