Model comparison
DeepSeek-V3.1 vs Gemma 2 9B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
DeepSeek
GoogleSummary
- They share 19 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 8 categories and Gemma 2 9B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.7 to 9.7.
Side by side
| DeepSeek-V3.1 | Gemma 2 9B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.8 | 25.9 |
| Released | 2025-08-21 | 2024-06-24 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Gemma 2 9B: 29.4 (#304)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Coding | 1417 | 1173 |
| WeirdML | 38.4% | — |
| BigCodeBench Instruct | — | 34.7% |
| LiveBench Coding | — | 22.5% |
| BigCodeBench Complete | — | 40.6% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Gemma 2 9B: 15.9 (#309)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1171 |
| Epoch Capabilities Index | 139.92 | 119.83 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| LiveBench Reasoning | — | 15.2% |
| DTBench | 82.7% | — |
| LiveBench Data Analysis | — | 36.4% |
| LMCA | 24.3% | — |
| ForecastBench | 58 | — |
| LiveBench | — | 28.7% |
| PIQA | — | 83.7% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Gemma 2 9B: 9.9 (#318)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Math | 1420 | 1183 |
| OTIS Mock AIME 2024-2025 | — | 0.6% |
| LiveBench Math | — | 19.8% |
| MATH Level 5 | — | 21% |
| GSM8K | — | 84.9% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Gemma 2 9B: 9.7 (#305)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Expert | 1405 | 1147 |
| GPQA Diamond | — | 27.5% |
| Vectara Hallucination Rate | 5.5% | — |
| BoolQ | — | 85.7% |
| MMLU | — | 72.1% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Gemma 2 9B: 36.6 (#238)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Non-English | 1400 | 1188 |
| LMArena Chinese | 1469 | 1185 |
| LMArena French | 1447 | 1190 |
| LMArena German | 1411 | 1186 |
| LMArena Japanese | 1378 | 1144 |
| LMArena Korean | 1337 | 1137 |
| LMArena Russian | 1405 | 1200 |
| LMArena Spanish | 1431 | 1200 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Gemma 2 9B: 57.6 (#269)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1178 |
| LiveBench Instruction Following | — | 52.6% |
Long Context Too close to call
DeepSeek-V3.1: 36.3 (#232), Gemma 2 9B: 36.3 (#233)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Longer Query | 1422 | 1197 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Gemma 2 9B: 32.1 (#281)
| Benchmark | DeepSeek-V3.1 | Gemma 2 9B |
|---|---|---|
| LMArena Text | 1420 | 1207 |
| LMArena Creative Writing | 1401 | 1206 |
| EQ-Bench Creative Writing | 1436 | 841 |
| LMArena Multi-Turn | 1408 | 1193 |
| LiveBench Language | — | 25.5% |
Frequently asked questions
Is DeepSeek-V3.1 better than Gemma 2 9B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.9 on the Noometry Index.
Is DeepSeek-V3.1 or Gemma 2 9B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 29.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Gemma 2 9B share?
19 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Gemma 2 9B has 35.