Model comparison
Gemma 2 27B vs Llama 2-70B
Gemma 2 27B is the stronger model overall, scoring 29.4 to 24.4 on the Noometry Index.
Last verified . 23 shared benchmarks.
Google
MetaSummary
- They share 23 benchmarks with published results for both. Gemma 2 27B scores higher in 8 categories and Llama 2-70B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 2 27B leads 44.2 to 32.3.
- The biggest single-benchmark swing is MATH Level 5: 27.9% for Gemma 2 27B and 3.3% for Llama 2-70B.
Side by side
| Gemma 2 27B | Llama 2-70B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 29.4 | 24.4 |
| Released | 2024-06-24 | 2023-07-18 |
| 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), Llama 2-70B: 31.4 (#286)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1211 | 1079 |
| BigCodeBench Instruct | 42.8% | — |
| LiveBench Coding | 36% | — |
| BigCodeBench Complete | 52.5% | — |
Reasoning Too close to call
Gemma 2 27B: 15.3 (#315), Llama 2-70B: 14.4 (#325)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1198 | 1073 |
| DTBench | 48% | 41.6% |
| Epoch Capabilities Index | 122.08 | 113.79 |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 47.9% | — |
| LMCA | 7.1% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| LiveBench | 38.2% | — |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math Gemma 2 27B leads
Gemma 2 27B: 10.7 (#311), Llama 2-70B: 8.1 (#326)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 0% |
| LMArena Math | 1212 | 1091 |
| MATH Level 5 | 27.9% | 3.3% |
| LiveBench Math | 26.5% | — |
| GSM8K | — | 69.6% |
Knowledge Gemma 2 27B leads
Gemma 2 27B: 19.0 (#280), Llama 2-70B: 7.4 (#310)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 36.5% | 26.3% |
| LMArena Expert | 1172 | 1039 |
| MMLU | 75.7% | 69.9% |
| Confabulations | 27.1% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual Gemma 2 27B leads
Gemma 2 27B: 38.6 (#226), Llama 2-70B: 27.7 (#274)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1217 | 1045 |
| LMArena Chinese | 1221 | 995 |
| LMArena French | 1247 | 1090 |
| LMArena German | 1209 | 1041 |
| LMArena Japanese | 1175 | 927 |
| LMArena Korean | 1174 | 964 |
| LMArena Russian | 1234 | 1083 |
| LMArena Spanish | 1228 | 1143 |
Instruction Following Gemma 2 27B leads
Gemma 2 27B: 60.5 (#249), Llama 2-70B: 54.9 (#278)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1206 | 1071 |
| LiveBench Instruction Following | 58.1% | — |
Long Context Gemma 2 27B leads
Gemma 2 27B: 37.3 (#218), Llama 2-70B: 32.3 (#270)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1231 | 1062 |
Writing & Preference Gemma 2 27B leads
Gemma 2 27B: 44.2 (#225), Llama 2-70B: 32.3 (#279)
| Benchmark | Gemma 2 27B | Llama 2-70B |
|---|---|---|
| LMArena Text | 1231 | 1115 |
| LMArena Creative Writing | 1241 | 1075 |
| LMArena Multi-Turn | 1224 | 1088 |
| LiveBench Language | 32.6% | — |
Frequently asked questions
Is Gemma 2 27B better than Llama 2-70B?
Gemma 2 27B is the stronger model overall, scoring 29.4 to 24.4 on the Noometry Index.
Is Gemma 2 27B or Llama 2-70B better for coding?
Gemma 2 27B scores higher on coding benchmarks: 34.1 versus 31.4 in the Noometry coding category.
How many benchmarks do Gemma 2 27B and Llama 2-70B share?
23 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Llama 2-70B has 35.