Model comparison
DeepSeek-V2.5 (Sep 2024) vs Ministral 8B
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 28.2 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 12 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 7 categories and Ministral 8B 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 12.6.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Ministral 8B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 37.6 | 28.2 |
| Released | 2024-09-06 | 2024-10-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.15 |
| Results tracked | 22 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Ministral 8B: 35.0 (#230)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Coding | 1309 | 1202 |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Ministral 8B: 16.4 (#148)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Ministral 8B: 18.4 (#281)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1191 |
| DTBench | — | 45.7% |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Ministral 8B: 25.7 (#267)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Math | 1288 | 1188 |
| MATH Level 5 | — | 14.9% |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Ministral 8B: 12.6 (#297)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Expert | 1266 | 1170 |
| GPQA Diamond | — | 27.1% |
| Vectara Hallucination Rate | — | 7.4% |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Ministral 8B: 35.1 (#247)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1273 | 1165 |
| LMArena Chinese | 1318 | 1193 |
| LMArena Russian | 1289 | 1195 |
| LMArena French | 1289 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Spanish | 1248 | — |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Ministral 8B: 60.5 (#250)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1280 | 1161 |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Ministral 8B: 36.7 (#227)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1301 | 1212 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Ministral 8B: 39.6 (#246)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Ministral 8B |
|---|---|---|
| LMArena Text | 1294 | 1191 |
| LMArena Creative Writing | 1285 | 1175 |
| LMArena Multi-Turn | 1297 | 1166 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Ministral 8B?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 28.2 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Ministral 8B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Ministral 8B share?
12 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Ministral 8B has 17.