Model comparison
DeepSeek LLM 67B vs Gemma 2 9B
DeepSeek LLM 67B and Gemma 2 9B score almost the same on the Noometry Index (24.8 vs 25.7), so choose on price, context window or the category you care about most.
Last verified . 14 shared benchmarks.
DeepSeek
GoogleSummary
- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 2 categories and Gemma 2 9B in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 2 9B leads 36.8 to 29.5.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 21% for Gemma 2 9B.
Side by side
| DeepSeek LLM 67B | Gemma 2 9B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 24.8 | 25.7 |
| Released | 2023-11-29 | 2024-06-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.3 (#273), Gemma 2 9B: 29.0 (#299)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| LMArena Coding | 1096 | 1174 |
| BigCodeBench Instruct | — | 34.7% |
| LiveBench Coding | — | 22.5% |
| BigCodeBench Complete | — | 40.6% |
Reasoning Too close to call
DeepSeek LLM 67B: 15.9 (#300), Gemma 2 9B: 15.3 (#306)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1172 |
| Epoch Capabilities Index | 110.54 | 119.78 |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 15.2% |
| LiveBench Data Analysis | — | 36.4% |
| LiveBench | — | 28.7% |
| PIQA | — | 83.7% |
Math Gemma 2 9B leads
DeepSeek LLM 67B: 8.1 (#318), Gemma 2 9B: 9.3 (#312)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 0.6% |
| LMArena Math | 1108 | 1183 |
| MATH Level 5 | 6.4% | 21% |
| LiveBench Math | — | 19.8% |
| GSM8K | — | 84.9% |
Knowledge Gemma 2 9B leads
DeepSeek LLM 67B: 7.1 (#308), Gemma 2 9B: 9.7 (#300)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| GPQA Diamond | 24.6% | 27.5% |
| LMArena Expert | — | 1147 |
| BoolQ | — | 85.7% |
| MMLU | — | 72.1% |
Multilingual Gemma 2 9B leads
DeepSeek LLM 67B: 29.5 (#262), Gemma 2 9B: 36.8 (#233)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| LMArena Non-English | 1073 | 1188 |
| LMArena Chinese | 1132 | 1185 |
| LMArena French | — | 1189 |
| LMArena German | — | 1186 |
| LMArena Japanese | — | 1144 |
| LMArena Korean | — | 1137 |
| LMArena Russian | — | 1200 |
| LMArena Spanish | — | 1200 |
Instruction Following Gemma 2 9B leads
DeepSeek LLM 67B: 55.7 (#272), Gemma 2 9B: 57.8 (#264)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1178 |
| LiveBench Instruction Following | — | 52.6% |
Long Context Gemma 2 9B leads
DeepSeek LLM 67B: 33.1 (#260), Gemma 2 9B: 36.3 (#228)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| LMArena Longer Query | 1092 | 1197 |
Writing & Preference Too close to call
DeepSeek LLM 67B: 31.8 (#277), Gemma 2 9B: 32.3 (#276)
| Benchmark | DeepSeek LLM 67B | Gemma 2 9B |
|---|---|---|
| LMArena Text | 1105 | 1207 |
| LMArena Creative Writing | 1067 | 1206 |
| LMArena Multi-Turn | 1082 | 1193 |
| EQ-Bench Creative Writing | — | 841 |
| LiveBench Language | — | 25.5% |
Frequently asked questions
Is DeepSeek LLM 67B better than Gemma 2 9B?
DeepSeek LLM 67B and Gemma 2 9B score almost the same on the Noometry Index (24.8 vs 25.7), so choose on price, context window or the category you care about most.
Is DeepSeek LLM 67B or Gemma 2 9B better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.3 versus 29.0 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Gemma 2 9B share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Gemma 2 9B has 35.