Model comparison
DeepSeek LLM 67B vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 24.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 13 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Mistral Medium in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 31.6.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 81.6% for Mistral Medium.
Side by side
| DeepSeek LLM 67B | Mistral Medium | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 24.9 | 36.3 |
| Released | 2023-11-29 | 2023-12-11 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $1.50 |
| Output $ / M tokens | — | $7.50 |
| Results tracked | 15 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
DeepSeek LLM 67B: 31.9 (#278), Mistral Medium: 34.2 (#243)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Coding | 1096 | 1434 |
| FrontierCode | — | 8% |
| SciCode | — | 40.2% |
| WeirdML | — | 43.7% |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Mistral Medium: 28.3 (#90)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
Reasoning Mistral Medium leads
DeepSeek LLM 67B: 16.5 (#304), Mistral Medium: 24.0 (#167)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1426 |
| Kagi LLM Benchmark | — | 50% |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| DTBench | — | 75.5% |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 110.5 | — |
Math Mistral Medium leads
DeepSeek LLM 67B: 8.7 (#324), Mistral Medium: 28.1 (#245)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 32.2% |
| LMArena Math | 1108 | 1408 |
| MATH Level 5 | 6.4% | 81.6% |
| ProofBench | — | 9% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Medium leads
DeepSeek LLM 67B: 7.0 (#313), Mistral Medium: 25.0 (#265)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| GPQA Diamond | 24.6% | 59.5% |
| Humanity's Last Exam | — | 4.5% |
| Vectara Hallucination Rate | — | 22.7% |
| LMArena Expert | — | 1408 |
Multimodal Not comparable
DeepSeek LLM 67B: —, Mistral Medium: 35.3 (#88)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Mistral Medium leads
DeepSeek LLM 67B: 29.4 (#267), Mistral Medium: 52.1 (#91)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1073 | 1408 |
| LMArena Chinese | 1132 | 1447 |
| LMArena French | — | 1459 |
| LMArena German | — | 1432 |
| LMArena Japanese | — | 1378 |
| LMArena Korean | — | 1380 |
| LMArena Russian | — | 1411 |
| LMArena Spanish | — | 1433 |
Instruction Following Mistral Medium leads
DeepSeek LLM 67B: 55.4 (#277), Mistral Medium: 73.7 (#116)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1079 | 1398 |
Long Context Mistral Medium leads
DeepSeek LLM 67B: 33.1 (#265), Mistral Medium: 42.9 (#114)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1092 | 1406 |
Writing & Preference Mistral Medium leads
DeepSeek LLM 67B: 31.6 (#282), Mistral Medium: 60.0 (#103)
| Benchmark | DeepSeek LLM 67B | Mistral Medium |
|---|---|---|
| LMArena Text | 1105 | 1424 |
| LMArena Creative Writing | 1067 | 1391 |
| LMArena Multi-Turn | 1082 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
Frequently asked questions
Is DeepSeek LLM 67B better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Mistral Medium share?
13 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Mistral Medium has 36.