Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemma 2B
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 29.6 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
GoogleSummary
- They share 13 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 7 categories and Gemma 2B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V2.5 (Sep 2024) leads 49.8 to 24.0.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemma 2B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 29.6 |
| Released | 2024-09-06 | 2024-02-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 23 |
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 2B: 29.4 (#305)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Coding | 1309 | 1010 |
| HumanEval+ | 83.5% | 20.7% |
| MBPP+ | 74.1% | 34.1% |
| 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 2B: 18.8 (#275)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Hard Prompts | 1289 | 989 |
| BIG-Bench Hard | — | 35.2% |
| Epoch Capabilities Index | — | 94.2 |
| HellaSwag | — | 71.4% |
| PIQA | — | 77.3% |
| WinoGrande | — | 65.4% |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemma 2B: 30.0 (#239)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Math | 1288 | 1009 |
| GSM8K | — | 17.7% |
Knowledge Not comparable
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemma 2B: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Expert | 1266 | — |
| ARC (AI2) Challenge | — | 42.1% |
| BoolQ | — | 69.4% |
| MMLU | — | 42.3% |
| TriviaQA | — | 53.2% |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemma 2B: 23.0 (#294)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Non-English | 1273 | 958 |
| LMArena Chinese | 1318 | 986 |
| LMArena Russian | 1289 | 937 |
| LMArena French | 1289 | — |
| 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 2B: 48.5 (#302)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Instruction Following | 1280 | 970 |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemma 2B: 29.9 (#291)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Longer Query | 1301 | 981 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemma 2B: 24.0 (#308)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemma 2B |
|---|---|---|
| LMArena Text | 1294 | 1002 |
| LMArena Creative Writing | 1285 | 987 |
| LMArena Multi-Turn | 1297 | 945 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemma 2B?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 29.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemma 2B better for coding?
DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.7 versus 29.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemma 2B share?
13 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemma 2B has 23.