Model comparison
Gemma 1.1 7b IT vs Llama 2-13B
Gemma 1.1 7b IT is the stronger model overall, scoring 30.7 to 29.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
Google
MetaSummary
- They share 17 benchmarks with published results for both. Gemma 1.1 7b IT scores higher in 7 categories and Llama 2-13B in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemma 1.1 7b IT leads 19.6 to 12.3.
Side by side
| Gemma 1.1 7b IT | Llama 2-13B | |
|---|---|---|
| Provider | Meta | |
| Noometry Index | 30.7 | 29.1 |
| Released | — | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 19 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemma 1.1 7b IT: 31.0 (#279), Llama 2-13B: 30.4 (#287)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1084 | 1062 |
| HumanEval+ | 35.4% | — |
| MBPP+ | 45% | — |
Reasoning Gemma 1.1 7b IT leads
Gemma 1.1 7b IT: 19.6 (#235), Llama 2-13B: 12.3 (#333)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Hard Prompts | 1071 | 1051 |
| Chess Puzzles | — | 0% |
| DTBench | — | 42.2% |
| BIG-Bench Hard | — | 58.2% |
| Epoch Capabilities Index | — | 105.8 |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math Too close to call
Gemma 1.1 7b IT: 29.5 (#217), Llama 2-13B: 28.7 (#226)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Math | 1107 | 1065 |
| GSM8K | — | 36.9% |
Knowledge Too close to call
Gemma 1.1 7b IT: 27.8 (#242), Llama 2-13B: 27.5 (#244)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1039 | 1030 |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| MMLU | — | 55.6% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
Gemma 1.1 7b IT: —, Llama 2-13B: —
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| ScienceQA | — | 55.8% |
Multilingual Gemma 1.1 7b IT leads
Gemma 1.1 7b IT: 28.3 (#268), Llama 2-13B: 26.7 (#274)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1052 | 1024 |
| LMArena Chinese | 1060 | 1001 |
| LMArena French | 1064 | 1043 |
| LMArena German | 1054 | 1009 |
| LMArena Japanese | 971 | 894 |
| LMArena Korean | 988 | 953 |
| LMArena Russian | 1046 | 1055 |
| LMArena Spanish | 1049 | 1086 |
Instruction Following Too close to call
Gemma 1.1 7b IT: 54.3 (#278), Llama 2-13B: 53.5 (#282)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1057 | 1046 |
Long Context Too close to call
Gemma 1.1 7b IT: 32.0 (#267), Llama 2-13B: 32.3 (#264)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1056 | 1064 |
Writing & Preference Too close to call
Gemma 1.1 7b IT: 30.6 (#283), Llama 2-13B: 30.0 (#284)
| Benchmark | Gemma 1.1 7b IT | Llama 2-13B |
|---|---|---|
| LMArena Text | 1094 | 1085 |
| LMArena Creative Writing | 1059 | 1047 |
| LMArena Multi-Turn | 1040 | 1050 |
Frequently asked questions
Is Gemma 1.1 7b IT better than Llama 2-13B?
Gemma 1.1 7b IT is the stronger model overall, scoring 30.7 to 29.1 on the Noometry Index.
Is Gemma 1.1 7b IT or Llama 2-13B better for coding?
They score almost the same on coding (31.0 vs 30.4); test both on your own repository before choosing.
How many benchmarks do Gemma 1.1 7b IT and Llama 2-13B share?
17 benchmarks have published results for both models. Gemma 1.1 7b IT has 19 scored results on Noometry and Llama 2-13B has 32.