Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Gemma 7B
Gemma 7B has enough public results to be ranked (#295); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 8 shared benchmarks.
DeepSeek
GoogleSummary
- They share 8 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Gemma 7B in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 39.8 to 29.9.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 39.7 | 29.5 |
| Released | 2024-05-07 | 2024-02-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 27 |
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): 39.8 (#141), Gemma 7B: 29.9 (#290)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | — |
| LMArena Coding | — | 1048 |
| BigCodeBench Complete | 59.4% | — |
| HumanEval+ | — | 28.7% |
| MBPP+ | — | 43.4% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 19.0 (#247)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| BIG-Bench Hard | 78.8% | 55.1% |
| Epoch Capabilities Index | 124.78 | 111.74 |
| HellaSwag | 87.1% | 82.2% |
| PIQA | 83.9% | 81.2% |
| WinoGrande | 86.3% | 79% |
| LMArena Hard Prompts | — | 1042 |
| Adversarial NLI | — | 48.7% |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 28.7 (#225)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| LMArena Math | — | 1066 |
| GSM8K | — | 46.4% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 26.8 (#248)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| ARC (AI2) Challenge | 92.2% | 78.3% |
| MMLU | 78.4% | 66.1% |
| TriviaQA | 80% | 72.3% |
| LMArena Expert | — | 1002 |
| BoolQ | — | 83.2% |
| OpenBookQA | — | 78.6% |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 25.3 (#282)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| LMArena Non-English | — | 999 |
| LMArena Chinese | — | 1035 |
| LMArena French | — | 1024 |
| LMArena Russian | — | 993 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 51.7 (#290)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| LMArena Instruction Following | — | 1017 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 31.1 (#277)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| LMArena Longer Query | — | 1022 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemma 7B: 27.2 (#297)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemma 7B |
|---|---|---|
| LMArena Text | — | 1056 |
| LMArena Creative Writing | — | 1024 |
| LMArena Multi-Turn | — | 963 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Gemma 7B?
Gemma 7B has enough public results to be ranked (#295); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Gemma 7B better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 39.8 versus 29.9 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Gemma 7B share?
8 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Gemma 7B has 27.