Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Gemini 1.0 Pro
Gemini 1.0 Pro has enough public results to be ranked (#332); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 2 shared benchmarks.
DeepSeek
GoogleSummary
- They share 2 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Gemini 1.0 Pro 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 32.2.
- DeepSeek-V2 (MoE-236B, May 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 40.3 | 27.3 |
| Released | 2024-05-07 | 2023-12-13 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 24 |
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), Gemini 1.0 Pro: 32.2 (#275)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| BigCodeBench Instruct | 48.9% | — |
| LMArena Coding | — | 1108 |
| BigCodeBench Complete | 59.4% | — |
| HumanEval+ | — | 55.5% |
| MBPP+ | — | 61.4% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 17.1 (#296)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| Epoch Capabilities Index | 124.77 | 117.04 |
| LMArena Hard Prompts | — | 1109 |
| DTBench | — | 45.9% |
| BIG-Bench Hard | 78.8% | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 9.3 (#321)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 1.1% |
| LMArena Math | — | 1132 |
| MATH Level 5 | — | 11.2% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 15.6 (#291)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| MMLU | 78.4% | 70% |
| GPQA Diamond | — | 34% |
| LMArena Expert | — | 1059 |
| ARC (AI2) Challenge | 92.2% | — |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 33.4 (#252)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Non-English | — | 1138 |
| LMArena Chinese | — | 1124 |
| LMArena French | — | 1145 |
| LMArena German | — | 1125 |
| LMArena Japanese | — | 1023 |
| LMArena Russian | — | 1186 |
| LMArena Spanish | — | 1119 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 57.6 (#267)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Instruction Following | — | 1114 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 34.3 (#249)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Longer Query | — | 1132 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 1.0 Pro: 36.0 (#264)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Text | — | 1149 |
| LMArena Creative Writing | — | 1131 |
| LMArena Multi-Turn | — | 1139 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Gemini 1.0 Pro?
Gemini 1.0 Pro has enough public results to be ranked (#332); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Gemini 1.0 Pro better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 32.2 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Gemini 1.0 Pro share?
2 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Gemini 1.0 Pro has 24.