Model comparison
Deepseek Coder v2 vs Gemma 2 9B
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 25.9 on the Noometry Index.
Last verified . 20 shared benchmarks.
DeepSeek
GoogleSummary
- They share 20 benchmarks with published results for both. Deepseek Coder v2 scores higher in 7 categories and Gemma 2 9B in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 9.9.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.7% for Deepseek Coder v2 and 40.6% for Gemma 2 9B.
Side by side
| Deepseek Coder v2 | Gemma 2 9B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 35.9 | 25.9 |
| Released | 2024-06-17 | 2024-06-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), Gemma 2 9B: 29.4 (#304)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| BigCodeBench Instruct | 48.2% | 34.7% |
| LMArena Coding | 1251 | 1173 |
| BigCodeBench Complete | 59.7% | 40.6% |
| LiveBench Coding | — | 22.5% |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Gemma 2 9B: 15.9 (#309)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1171 |
| LiveBench Reasoning | — | 15.2% |
| LiveBench Data Analysis | — | 36.4% |
| Epoch Capabilities Index | — | 119.83 |
| LiveBench | — | 28.7% |
| PIQA | — | 83.7% |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Gemma 2 9B: 9.9 (#318)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Math | 1241 | 1183 |
| GSM8K | 94.5% | 84.9% |
| OTIS Mock AIME 2024-2025 | — | 0.6% |
| LiveBench Math | — | 19.8% |
| MATH Level 5 | — | 21% |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Gemma 2 9B: 9.7 (#305)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Expert | 1181 | 1147 |
| GPQA Diamond | — | 27.5% |
| ARC (AI2) Challenge | 64.3% | — |
| BoolQ | — | 85.7% |
| MMLU | — | 72.1% |
Multilingual Too close to call
Deepseek Coder v2: 36.3 (#240), Gemma 2 9B: 36.6 (#238)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Non-English | 1182 | 1188 |
| LMArena Chinese | 1201 | 1185 |
| LMArena French | 1185 | 1190 |
| LMArena German | 1164 | 1186 |
| LMArena Japanese | 1126 | 1144 |
| LMArena Korean | 1104 | 1137 |
| LMArena Russian | 1188 | 1200 |
| LMArena Spanish | 1153 | 1200 |
Instruction Following Deepseek Coder v2 leads
Deepseek Coder v2: 61.7 (#242), Gemma 2 9B: 57.6 (#269)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Instruction Following | 1180 | 1178 |
| LiveBench Instruction Following | — | 52.6% |
Long Context Too close to call
Deepseek Coder v2: 37.0 (#224), Gemma 2 9B: 36.3 (#233)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Longer Query | 1219 | 1197 |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.2 (#253), Gemma 2 9B: 32.1 (#281)
| Benchmark | Deepseek Coder v2 | Gemma 2 9B |
|---|---|---|
| LMArena Text | 1191 | 1207 |
| LMArena Creative Writing | 1120 | 1206 |
| LMArena Multi-Turn | 1177 | 1193 |
| EQ-Bench Creative Writing | — | 841 |
| LiveBench Language | — | 25.5% |
Frequently asked questions
Is Deepseek Coder v2 better than Gemma 2 9B?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 25.9 on the Noometry Index.
Is Deepseek Coder v2 or Gemma 2 9B better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 29.4 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Gemma 2 9B share?
20 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Gemma 2 9B has 35.