Model comparison
Gemma 7B vs Llama 2-13B
Gemma 7B and Llama 2-13B score almost the same on the Noometry Index (30.0 vs 29.6), 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 2 categories and Llama 2-13B in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemma 7B leads 19.9 to 12.8.
Side by side
| Gemma 7B | Llama 2-13B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 30.0 | 29.6 |
| Released | 2024-02-21 | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 27 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemma 7B: 30.5 (#294), Llama 2-13B: 30.9 (#291)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1048 | 1062 |
| HumanEval+ | 28.7% | — |
| MBPP+ | 43.4% | — |
Reasoning Gemma 7B leads
Gemma 7B: 19.9 (#249), Llama 2-13B: 12.8 (#337)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1051 |
| BIG-Bench Hard | 55.1% | 58.2% |
| Epoch Capabilities Index | 111.99 | 106.17 |
| HellaSwag | 82.2% | 80.7% |
| PIQA | 81.2% | 80.8% |
| WinoGrande | 79% | 72.8% |
| Chess Puzzles | — | 0% |
| DTBench | — | 42.2% |
| Adversarial NLI | 48.7% | — |
| LAMBADA | — | 76.5% |
Math Too close to call
Gemma 7B: 31.2 (#228), Llama 2-13B: 31.1 (#229)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Math | 1066 | 1065 |
| GSM8K | 46.4% | 36.9% |
Knowledge Too close to call
Gemma 7B: 27.3 (#252), Llama 2-13B: 28.1 (#249)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1001 | 1030 |
| ARC (AI2) Challenge | 78.3% | 60.3% |
| BoolQ | 83.2% | 82.4% |
| MMLU | 66.1% | 55.6% |
| OpenBookQA | 78.6% | 57% |
| TriviaQA | 72.3% | 79.6% |
Multimodal Not comparable
Gemma 7B: —, Llama 2-13B: —
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| ScienceQA | — | 55.8% |
Multilingual Llama 2-13B leads
Gemma 7B: 25.1 (#287), Llama 2-13B: 26.5 (#279)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 999 | 1024 |
| LMArena Chinese | 1035 | 1001 |
| LMArena French | 1025 | 1044 |
| LMArena Russian | 993 | 1055 |
| LMArena German | — | 1009 |
| LMArena Japanese | — | 894 |
| LMArena Korean | — | 953 |
| LMArena Spanish | — | 1087 |
Instruction Following Llama 2-13B leads
Gemma 7B: 51.5 (#295), Llama 2-13B: 53.3 (#287)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1017 | 1045 |
Long Context Llama 2-13B leads
Gemma 7B: 31.1 (#282), Llama 2-13B: 32.3 (#269)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1022 | 1064 |
Writing & Preference Llama 2-13B leads
Gemma 7B: 27.1 (#302), Llama 2-13B: 29.8 (#289)
| Benchmark | Gemma 7B | Llama 2-13B |
|---|---|---|
| LMArena Text | 1056 | 1084 |
| LMArena Creative Writing | 1024 | 1047 |
| LMArena Multi-Turn | 963 | 1050 |
Frequently asked questions
Is Gemma 7B better than Llama 2-13B?
Gemma 7B and Llama 2-13B score almost the same on the Noometry Index (30.0 vs 29.6), so choose on price, context window or the category you care about most.
Is Gemma 7B or Llama 2-13B better for coding?
They score almost the same on coding (30.5 vs 30.9); test both on your own repository before choosing.
How many benchmarks do Gemma 7B and Llama 2-13B share?
24 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Llama 2-13B has 32.