Model comparison
Llama 2-13B vs Llama 2-7B
Llama 2-13B and Llama 2-7B score almost the same on the Noometry Index (29.6 vs 29.1), so choose on price, context window or the category you care about most.
Last verified . 29 shared benchmarks.
MetaSummary
- They share 29 benchmarks with published results for both. Llama 2-13B scores higher in 6 categories and Llama 2-7B in 2 categories; 6 gaps are clear of the uncertainty.
Side by side
| Llama 2-13B | Llama 2-7B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 29.6 | 29.1 |
| Released | 2023-07-18 | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 32 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-13B leads
Llama 2-13B: 30.9 (#291), Llama 2-7B: 29.2 (#307)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1062 | 1002 |
Reasoning Llama 2-7B leads
Llama 2-13B: 12.8 (#337), Llama 2-7B: 15.7 (#312)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1051 | 1009 |
| BIG-Bench Hard | 58.2% | 39.2% |
| Epoch Capabilities Index | 106.17 | 99.06 |
| HellaSwag | 80.7% | 77.2% |
| LAMBADA | 76.5% | 73.3% |
| PIQA | 80.8% | 78.8% |
| WinoGrande | 72.8% | 69.2% |
| DTBench | 42.2% | — |
Math Too close to call
Llama 2-13B: 31.1 (#229), Llama 2-7B: 30.7 (#233)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Math | 1065 | 1042 |
| GSM8K | 36.9% | 16.7% |
Knowledge Too close to call
Llama 2-13B: 28.1 (#249), Llama 2-7B: 28.2 (#248)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Expert | 1030 | 1036 |
| ARC (AI2) Challenge | 60.3% | 45.9% |
| BoolQ | 82.4% | 77.9% |
| MMLU | 55.6% | 45.8% |
| OpenBookQA | 57% | 58.6% |
| TriviaQA | 79.6% | 73.7% |
Multimodal Not comparable
Llama 2-13B: —, Llama 2-7B: —
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| ScienceQA | 55.8% | 43.1% |
Multilingual Llama 2-13B leads
Llama 2-13B: 26.5 (#279), Llama 2-7B: 23.8 (#293)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 1024 | 973 |
| LMArena Chinese | 1001 | 973 |
| LMArena French | 1044 | 970 |
| LMArena German | 1009 | 978 |
| LMArena Russian | 1055 | 995 |
| LMArena Spanish | 1087 | 1007 |
| LMArena Japanese | 894 | — |
| LMArena Korean | 953 | — |
Instruction Following Llama 2-13B leads
Llama 2-13B: 53.3 (#287), Llama 2-7B: 50.8 (#298)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1045 | 1006 |
Long Context Llama 2-13B leads
Llama 2-13B: 32.3 (#269), Llama 2-7B: 30.4 (#287)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1064 | 999 |
Writing & Preference Llama 2-13B leads
Llama 2-13B: 29.8 (#289), Llama 2-7B: 28.0 (#298)
| Benchmark | Llama 2-13B | Llama 2-7B |
|---|---|---|
| LMArena Text | 1084 | 1053 |
| LMArena Creative Writing | 1047 | 1033 |
| LMArena Multi-Turn | 1050 | 1029 |
Frequently asked questions
Is Llama 2-13B better than Llama 2-7B?
Llama 2-13B and Llama 2-7B score almost the same on the Noometry Index (29.6 vs 29.1), so choose on price, context window or the category you care about most.
Is Llama 2-13B or Llama 2-7B better for coding?
Llama 2-13B scores higher on coding benchmarks: 30.9 versus 29.2 in the Noometry coding category.
How many benchmarks do Llama 2-13B and Llama 2-7B share?
29 benchmarks have published results for both models. Llama 2-13B has 32 scored results on Noometry and Llama 2-7B has 29.