Model comparison
DeepSeek LLM 67B vs Ministral 8B
Ministral 8B is the stronger model overall, scoring 28.2 to 24.9 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 12 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Ministral 8B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Ministral 8B leads 25.7 to 8.7.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 14.9% for Ministral 8B.
Side by side
| DeepSeek LLM 67B | Ministral 8B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 24.9 | 28.2 |
| Released | 2023-11-29 | 2024-10-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.15 |
| Results tracked | 15 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
DeepSeek LLM 67B: 31.9 (#278), Ministral 8B: 35.0 (#230)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Coding | 1096 | 1202 |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Ministral 8B: 16.4 (#148)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
Reasoning Ministral 8B leads
DeepSeek LLM 67B: 16.5 (#304), Ministral 8B: 18.4 (#281)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1191 |
| Chess Puzzles | 0% | — |
| DTBench | — | 45.7% |
| Epoch Capabilities Index | 110.5 | — |
Math Ministral 8B leads
DeepSeek LLM 67B: 8.7 (#324), Ministral 8B: 25.7 (#267)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Math | 1108 | 1188 |
| MATH Level 5 | 6.4% | 14.9% |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
Knowledge Ministral 8B leads
DeepSeek LLM 67B: 7.0 (#313), Ministral 8B: 12.6 (#297)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| GPQA Diamond | 24.6% | 27.1% |
| Vectara Hallucination Rate | — | 7.4% |
| LMArena Expert | — | 1170 |
Multilingual Ministral 8B leads
DeepSeek LLM 67B: 29.4 (#267), Ministral 8B: 35.1 (#247)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1073 | 1165 |
| LMArena Chinese | 1132 | 1193 |
| LMArena Russian | — | 1195 |
Instruction Following Ministral 8B leads
DeepSeek LLM 67B: 55.4 (#277), Ministral 8B: 60.5 (#250)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1161 |
Long Context Ministral 8B leads
DeepSeek LLM 67B: 33.1 (#265), Ministral 8B: 36.7 (#227)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1092 | 1212 |
Writing & Preference Ministral 8B leads
DeepSeek LLM 67B: 31.6 (#282), Ministral 8B: 39.6 (#246)
| Benchmark | DeepSeek LLM 67B | Ministral 8B |
|---|---|---|
| LMArena Text | 1105 | 1191 |
| LMArena Creative Writing | 1067 | 1175 |
| LMArena Multi-Turn | 1082 | 1166 |
Frequently asked questions
Is DeepSeek LLM 67B better than Ministral 8B?
Ministral 8B is the stronger model overall, scoring 28.2 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Ministral 8B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Ministral 8B share?
12 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Ministral 8B has 17.