Model comparison
Gemma 7B vs Llama 13b
Gemma 7B is the stronger model overall, scoring 29.5 to 23.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
Google
MetaSummary
- They share 19 benchmarks with published results for both. Gemma 7B scores higher in 6 categories and Llama 13b in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Gemma 7B leads 51.7 to 36.9.
Side by side
| Gemma 7B | Llama 13b | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 29.5 | 23.9 |
| Released | 2024-02-21 | 2023-02-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 27 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 7B leads
Gemma 7B: 29.9 (#290), Llama 13b: 20.9 (#332)
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Coding | 1048 | 683 |
| HumanEval+ | 28.7% | — |
| MBPP+ | 43.4% | — |
Reasoning Gemma 7B leads
Gemma 7B: 19.0 (#247), Llama 13b: 13.4 (#327)
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Hard Prompts | 1042 | 728 |
| BIG-Bench Hard | 55.1% | 37.9% |
| Epoch Capabilities Index | 111.74 | 100.09 |
| HellaSwag | 82.2% | 79.2% |
| PIQA | 81.2% | 80.1% |
| WinoGrande | 79% | 73% |
| Adversarial NLI | 48.7% | — |
| LAMBADA | — | 75.2% |
Math Gemma 7B leads
Gemma 7B: 28.7 (#225), Llama 13b: 24.4 (#259)
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Math | 1066 | 838 |
| GSM8K | 46.4% | 20.6% |
Knowledge Not comparable
Gemma 7B: 26.8 (#248), Llama 13b: —
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| ARC (AI2) Challenge | 78.3% | 52.7% |
| BoolQ | 83.2% | 78.7% |
| MMLU | 66.1% | 47.7% |
| OpenBookQA | 78.6% | 56.4% |
| TriviaQA | 72.3% | 77.9% |
| LMArena Expert | 1002 | — |
Multimodal Not comparable
Gemma 7B: —, Llama 13b: —
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| ScienceQA | — | 43.3% |
Multilingual Gemma 7B leads
Gemma 7B: 25.3 (#282), Llama 13b: 16.8 (#292)
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Non-English | 999 | 819 |
| LMArena Chinese | 1035 | — |
| LMArena French | 1024 | — |
| LMArena Russian | 993 | — |
Instruction Following Gemma 7B leads
Gemma 7B: 51.7 (#290), Llama 13b: 36.9 (#300)
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Instruction Following | 1017 | 781 |
Long Context Not comparable
Gemma 7B: 31.1 (#277), Llama 13b: —
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Longer Query | 1022 | — |
Writing & Preference Gemma 7B leads
Gemma 7B: 27.2 (#297), Llama 13b: 13.9 (#307)
| Benchmark | Gemma 7B | Llama 13b |
|---|---|---|
| LMArena Text | 1056 | 834 |
| LMArena Creative Writing | 1024 | 794 |
| LMArena Multi-Turn | 963 | 753 |
Frequently asked questions
Is Gemma 7B better than Llama 13b?
Gemma 7B is the stronger model overall, scoring 29.5 to 23.9 on the Noometry Index.
Is Gemma 7B or Llama 13b better for coding?
Gemma 7B scores higher on coding benchmarks: 29.9 versus 20.9 in the Noometry coding category.
How many benchmarks do Gemma 7B and Llama 13b share?
19 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Llama 13b has 21.