Model comparison
Gemma 7B vs Llama 2-7B
Gemma 7B and Llama 2-7B score almost the same on the Noometry Index (30.0 vs 29.1), so choose on price, context window or the category you care about most.
Last verified . 24 shared benchmarks.
Google
MetaSummary
- They share 24 benchmarks with published results for both. Gemma 7B scores higher in 6 categories and Llama 2-7B in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemma 7B leads 19.9 to 15.7.
Side by side
| Gemma 7B | Llama 2-7B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 30.0 | 29.1 |
| Released | 2024-02-21 | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 27 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 7B leads
Gemma 7B: 30.5 (#294), Llama 2-7B: 29.2 (#307)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1048 | 1002 |
| HumanEval+ | 28.7% | — |
| MBPP+ | 43.4% | — |
Reasoning Gemma 7B leads
Gemma 7B: 19.9 (#249), Llama 2-7B: 15.7 (#312)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1009 |
| BIG-Bench Hard | 55.1% | 39.2% |
| Epoch Capabilities Index | 111.99 | 99.06 |
| HellaSwag | 82.2% | 77.2% |
| PIQA | 81.2% | 78.8% |
| WinoGrande | 79% | 69.2% |
| Chess Puzzles | — | 0% |
| Adversarial NLI | 48.7% | — |
| LAMBADA | — | 73.3% |
Math Too close to call
Gemma 7B: 31.2 (#228), Llama 2-7B: 30.7 (#233)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Math | 1066 | 1042 |
| GSM8K | 46.4% | 16.7% |
Knowledge Too close to call
Gemma 7B: 27.3 (#252), Llama 2-7B: 28.2 (#248)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Expert | 1001 | 1036 |
| ARC (AI2) Challenge | 78.3% | 45.9% |
| BoolQ | 83.2% | 77.9% |
| MMLU | 66.1% | 45.8% |
| OpenBookQA | 78.6% | 58.6% |
| TriviaQA | 72.3% | 73.7% |
Multimodal Not comparable
Gemma 7B: —, Llama 2-7B: —
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| ScienceQA | — | 43.1% |
Multilingual Gemma 7B leads
Gemma 7B: 25.1 (#287), Llama 2-7B: 23.8 (#293)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 999 | 973 |
| LMArena Chinese | 1035 | 973 |
| LMArena French | 1025 | 970 |
| LMArena Russian | 993 | 995 |
| LMArena German | — | 978 |
| LMArena Spanish | — | 1007 |
Instruction Following Too close to call
Gemma 7B: 51.5 (#295), Llama 2-7B: 50.8 (#298)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1017 | 1006 |
Long Context Too close to call
Gemma 7B: 31.1 (#282), Llama 2-7B: 30.4 (#287)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1022 | 999 |
Writing & Preference Too close to call
Gemma 7B: 27.1 (#302), Llama 2-7B: 28.0 (#298)
| Benchmark | Gemma 7B | Llama 2-7B |
|---|---|---|
| LMArena Text | 1056 | 1053 |
| LMArena Creative Writing | 1024 | 1033 |
| LMArena Multi-Turn | 963 | 1029 |
Frequently asked questions
Is Gemma 7B better than Llama 2-7B?
Gemma 7B and Llama 2-7B score almost the same on the Noometry Index (30.0 vs 29.1), so choose on price, context window or the category you care about most.
Is Gemma 7B or Llama 2-7B better for coding?
Gemma 7B scores higher on coding benchmarks: 30.5 versus 29.2 in the Noometry coding category.
How many benchmarks do Gemma 7B and Llama 2-7B share?
24 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Llama 2-7B has 29.