Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemma 7B
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 30.0 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeek
GoogleSummary
- They share 15 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 8 categories and Gemma 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V2.5 (Sep 2024) leads 49.8 to 27.1.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemma 7B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 30.0 |
| Released | 2024-09-06 | 2024-02-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Gemma 7B: 30.5 (#294)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Coding | 1309 | 1048 |
| HumanEval+ | 83.5% | 28.7% |
| MBPP+ | 74.1% | 43.4% |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Gemma 7B: 19.9 (#249)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1042 |
| Adversarial NLI | — | 48.7% |
| BIG-Bench Hard | — | 55.1% |
| Epoch Capabilities Index | — | 111.99 |
| HellaSwag | — | 82.2% |
| PIQA | — | 81.2% |
| WinoGrande | — | 79% |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemma 7B: 31.2 (#228)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Math | 1288 | 1066 |
| GSM8K | — | 46.4% |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemma 7B: 27.3 (#252)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Expert | 1266 | 1001 |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 83.2% |
| MMLU | — | 66.1% |
| OpenBookQA | — | 78.6% |
| TriviaQA | — | 72.3% |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemma 7B: 25.1 (#287)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Non-English | 1273 | 999 |
| LMArena Chinese | 1318 | 1035 |
| LMArena French | 1289 | 1025 |
| LMArena Russian | 1289 | 993 |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Spanish | 1248 | — |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Gemma 7B: 51.5 (#295)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Instruction Following | 1280 | 1017 |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemma 7B: 31.1 (#282)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Longer Query | 1301 | 1022 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemma 7B: 27.1 (#302)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 7B |
|---|---|---|
| LMArena Text | 1294 | 1056 |
| LMArena Creative Writing | 1285 | 1024 |
| LMArena Multi-Turn | 1297 | 963 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemma 7B?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 30.0 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemma 7B better for coding?
DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.7 versus 30.5 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemma 7B share?
15 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemma 7B has 27.