Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Gemma 2 9B
Gemma 2 9B has enough public results to be ranked (#341); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 5 shared benchmarks.
DeepSeek
GoogleSummary
- They share 5 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Gemma 2 9B 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 29.4.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.4% for DeepSeek-V2 (MoE-236B, May 2024) and 40.6% for Gemma 2 9B.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 40.3 | 25.9 |
| Released | 2024-05-07 | 2024-06-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 35 |
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), Gemma 2 9B: 29.4 (#304)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | 34.7% |
| BigCodeBench Complete | 59.4% | 40.6% |
| LiveBench Coding | — | 22.5% |
| LMArena Coding | — | 1173 |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 15.9 (#309)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| Epoch Capabilities Index | 124.77 | 119.83 |
| PIQA | 83.9% | 83.7% |
| LiveBench Reasoning | — | 15.2% |
| LMArena Hard Prompts | — | 1171 |
| LiveBench Data Analysis | — | 36.4% |
| BIG-Bench Hard | 78.8% | — |
| HellaSwag | 87.1% | — |
| LiveBench | — | 28.7% |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 9.9 (#318)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 0.6% |
| LiveBench Math | — | 19.8% |
| LMArena Math | — | 1183 |
| MATH Level 5 | — | 21% |
| GSM8K | — | 84.9% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 9.7 (#305)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| MMLU | 78.4% | 72.1% |
| GPQA Diamond | — | 27.5% |
| LMArena Expert | — | 1147 |
| ARC (AI2) Challenge | 92.2% | — |
| BoolQ | — | 85.7% |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 36.6 (#238)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| LMArena Non-English | — | 1188 |
| LMArena Chinese | — | 1185 |
| LMArena French | — | 1190 |
| LMArena German | — | 1186 |
| LMArena Japanese | — | 1144 |
| LMArena Korean | — | 1137 |
| LMArena Russian | — | 1200 |
| LMArena Spanish | — | 1200 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 57.6 (#269)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| LiveBench Instruction Following | — | 52.6% |
| LMArena Instruction Following | — | 1178 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 36.3 (#233)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| LMArena Longer Query | — | 1197 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 2 9B: 32.1 (#281)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 2 9B |
|---|---|---|
| LMArena Text | — | 1207 |
| LMArena Creative Writing | — | 1206 |
| EQ-Bench Creative Writing | — | 841 |
| LMArena Multi-Turn | — | 1193 |
| LiveBench Language | — | 25.5% |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Gemma 2 9B?
Gemma 2 9B has enough public results to be ranked (#341); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Gemma 2 9B better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 29.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Gemma 2 9B share?
5 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Gemma 2 9B has 35.