Model comparison
Gemma 2 27B vs Gemma 2 9B
Gemma 2 27B is the stronger model overall, scoring 29.4 to 25.9 on the Noometry Index.
Last verified . 31 shared benchmarks.
GoogleSummary
- They share 31 benchmarks with published results for both. Gemma 2 27B scores higher in 7 categories and Gemma 2 9B in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 2 27B leads 44.2 to 32.1.
- The biggest single-benchmark swing is LiveBench Coding: 36% for Gemma 2 27B and 22.5% for Gemma 2 9B.
Side by side
| Gemma 2 27B | Gemma 2 9B | |
|---|---|---|
| Provider | ||
| Noometry Index | 29.4 | 25.9 |
| Released | 2024-06-24 | 2024-06-24 |
| Weights | Open | Open |
| Context window | 8K | — |
| Max output | 2K | — |
| Input $ / M tokens | $0.65 | — |
| Output $ / M tokens | $0.65 | — |
| Results tracked | 34 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 27B leads
Gemma 2 27B: 34.1 (#246), Gemma 2 9B: 29.4 (#304)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| BigCodeBench Instruct | 42.8% | 34.7% |
| LiveBench Coding | 36% | 22.5% |
| LMArena Coding | 1211 | 1173 |
| BigCodeBench Complete | 52.5% | 40.6% |
Reasoning Too close to call
Gemma 2 27B: 15.3 (#315), Gemma 2 9B: 15.9 (#309)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| LiveBench Reasoning | 28.1% | 15.2% |
| LMArena Hard Prompts | 1198 | 1171 |
| LiveBench Data Analysis | 47.9% | 36.4% |
| Epoch Capabilities Index | 122.08 | 119.83 |
| LiveBench | 38.2% | 28.7% |
| DTBench | 48% | — |
| LMCA | 7.1% | — |
| PIQA | — | 83.7% |
Math Too close to call
Gemma 2 27B: 10.7 (#311), Gemma 2 9B: 9.9 (#318)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 0.6% |
| LiveBench Math | 26.5% | 19.8% |
| LMArena Math | 1212 | 1183 |
| MATH Level 5 | 27.9% | 21% |
| GSM8K | — | 84.9% |
Knowledge Gemma 2 27B leads
Gemma 2 27B: 19.0 (#280), Gemma 2 9B: 9.7 (#305)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| GPQA Diamond | 36.5% | 27.5% |
| LMArena Expert | 1172 | 1147 |
| MMLU | 75.7% | 72.1% |
| Confabulations | 27.1% | — |
| BoolQ | — | 85.7% |
Multilingual Gemma 2 27B leads
Gemma 2 27B: 38.6 (#226), Gemma 2 9B: 36.6 (#238)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| LMArena Non-English | 1217 | 1188 |
| LMArena Chinese | 1221 | 1185 |
| LMArena French | 1247 | 1190 |
| LMArena German | 1209 | 1186 |
| LMArena Japanese | 1175 | 1144 |
| LMArena Korean | 1174 | 1137 |
| LMArena Russian | 1234 | 1200 |
| LMArena Spanish | 1228 | 1200 |
Instruction Following Gemma 2 27B leads
Gemma 2 27B: 60.5 (#249), Gemma 2 9B: 57.6 (#269)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| LiveBench Instruction Following | 58.1% | 52.6% |
| LMArena Instruction Following | 1206 | 1178 |
Long Context Gemma 2 27B leads
Gemma 2 27B: 37.3 (#218), Gemma 2 9B: 36.3 (#233)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| LMArena Longer Query | 1231 | 1197 |
Writing & Preference Gemma 2 27B leads
Gemma 2 27B: 44.2 (#225), Gemma 2 9B: 32.1 (#281)
| Benchmark | Gemma 2 27B | Gemma 2 9B |
|---|---|---|
| LMArena Text | 1231 | 1207 |
| LMArena Creative Writing | 1241 | 1206 |
| LMArena Multi-Turn | 1224 | 1193 |
| LiveBench Language | 32.6% | 25.5% |
| EQ-Bench Creative Writing | — | 841 |
Frequently asked questions
Is Gemma 2 27B better than Gemma 2 9B?
Gemma 2 27B is the stronger model overall, scoring 29.4 to 25.9 on the Noometry Index.
Is Gemma 2 27B or Gemma 2 9B better for coding?
Gemma 2 27B scores higher on coding benchmarks: 34.1 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2 27B and Gemma 2 9B share?
31 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Gemma 2 9B has 35.