Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Mixtral 8x7B
Mixtral 8x7B has enough public results to be ranked (#334); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 7 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 7 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Mixtral 8x7B in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 40.4 to 32.8.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 40.3 | 27.1 |
| Released | 2024-05-07 | 2023-12-11 |
| Weights | Open | Open |
| Context window | — | 32K |
| Max output | — | 32K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $0.70 |
| Results tracked | 10 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2 (MoE-236B, May 2024) leads
DeepSeek-V2 (MoE-236B, May 2024): 40.4 (#139), Mixtral 8x7B: 32.8 (#269)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | — |
| LMArena Coding | — | 1126 |
| BigCodeBench Complete | 59.4% | — |
| HumanEval+ | — | 39.6% |
| MBPP+ | — | 49.7% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 18.2 (#285)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| Epoch Capabilities Index | 124.77 | 118.47 |
| HellaSwag | 87.1% | 86.7% |
| PIQA | 83.9% | 83.6% |
| WinoGrande | 86.3% | 77.2% |
| LMArena Hard Prompts | — | 1115 |
| DTBench | — | 49.6% |
| Adversarial NLI | — | 55.2% |
| BIG-Bench Hard | 78.8% | — |
| ForecastBench | — | 56.3 |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 18.8 (#289)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| Omni-MATH | — | 10.5% |
| LMArena Math | — | 1147 |
| MATH Level 5 | — | 10% |
| GSM8K | — | 74.4% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 11.0 (#301)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| ARC (AI2) Challenge | 92.2% | 87.3% |
| MMLU | 78.4% | 70.6% |
| TriviaQA | 80% | 82.2% |
| GPQA Diamond | — | 30.6% |
| MMLU-Pro | — | 33.5% |
| GPQA (HELM) | — | 29.6% |
| LMArena Expert | — | 1088 |
| OpenBookQA | — | 85.8% |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 29.6 (#266)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | — | 1077 |
| LMArena Chinese | — | 1055 |
| LMArena French | — | 1166 |
| LMArena German | — | 1114 |
| LMArena Japanese | — | 931 |
| LMArena Korean | — | 968 |
| LMArena Russian | — | 1090 |
| LMArena Spanish | — | 1111 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 51.0 (#297)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| IFEval | — | 57.5% |
| LMArena Instruction Following | — | 1109 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 33.4 (#260)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | — | 1103 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Mixtral 8x7B: 34.2 (#270)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Mixtral 8x7B |
|---|---|---|
| LMArena Text | — | 1132 |
| LMArena Creative Writing | — | 1109 |
| WildBench | — | 67.3% |
| LMArena Multi-Turn | — | 1115 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Mixtral 8x7B?
Mixtral 8x7B has enough public results to be ranked (#334); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Mixtral 8x7B better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 32.8 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Mixtral 8x7B share?
7 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Mixtral 8x7B has 38.