Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Ministral 3B
Ministral 3B has enough public results to be ranked (#338); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 1 benchmark with published results for both.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Ministral 3B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 40.3 | 26.2 |
| Released | 2024-05-07 | 2024-10-01 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.10 |
| Results tracked | 10 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
DeepSeek-V2 (MoE-236B, May 2024): 40.4 (#139), Ministral 3B: —
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Ministral 3B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | — |
| BigCodeBench Complete | 59.4% | — |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Ministral 3B: 18.4 (#282)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Ministral 3B |
|---|---|---|
| Epoch Capabilities Index | 124.77 | 118.1 |
| DTBench | — | 51.7% |
| LMCA | — | 5.5% |
| BIG-Bench Hard | 78.8% | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Ministral 3B: 26.6 (#258)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Ministral 3B |
|---|---|---|
| MATH Level 5 | — | 14.4% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Ministral 3B: 10.4 (#302)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Ministral 3B |
|---|---|---|
| GPQA Diamond | — | 25.3% |
| Vectara Hallucination Rate | — | 7.3% |
| ARC (AI2) Challenge | 92.2% | — |
| MMLU | 78.4% | — |
| TriviaQA | 80% | — |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Ministral 3B?
Ministral 3B has enough public results to be ranked (#338); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Ministral 3B share?
1 benchmark has published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Ministral 3B has 6.