Model comparison
Codellama 70b Instruct vs Llama 2-7B
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 29.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
MetaSummary
- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 5 categories and Llama 2-7B in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where Codellama 70b Instruct leads 37.6 to 29.2.
Side by side
| Codellama 70b Instruct | Llama 2-7B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 33.7 | 29.1 |
| Released | — | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 7 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Codellama 70b Instruct leads
Codellama 70b Instruct: 37.6 (#193), Llama 2-7B: 29.2 (#307)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1002 |
| BigCodeBench Complete | 49.6% | — |
| HumanEval+ | 65.9% | — |
Reasoning Codellama 70b Instruct leads
Codellama 70b Instruct: 20.1 (#242), Llama 2-7B: 15.7 (#312)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1009 |
| Chess Puzzles | — | 0% |
| BIG-Bench Hard | — | 39.2% |
| Epoch Capabilities Index | — | 99.06 |
| HellaSwag | — | 77.2% |
| LAMBADA | — | 73.3% |
| PIQA | — | 78.8% |
| WinoGrande | — | 69.2% |
Math Not comparable
Codellama 70b Instruct: —, Llama 2-7B: 30.7 (#233)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Math | — | 1042 |
| GSM8K | — | 16.7% |
Knowledge Not comparable
Codellama 70b Instruct: —, Llama 2-7B: 28.2 (#248)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Expert | — | 1036 |
| ARC (AI2) Challenge | — | 45.9% |
| BoolQ | — | 77.9% |
| MMLU | — | 45.8% |
| OpenBookQA | — | 58.6% |
| TriviaQA | — | 73.7% |
Multimodal Not comparable
Codellama 70b Instruct: —, Llama 2-7B: —
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| ScienceQA | — | 43.1% |
Multilingual Codellama 70b Instruct leads
Codellama 70b Instruct: 24.8 (#288), Llama 2-7B: 23.8 (#293)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 992 | 973 |
| LMArena Chinese | — | 973 |
| LMArena French | — | 970 |
| LMArena German | — | 978 |
| LMArena Russian | — | 995 |
| LMArena Spanish | — | 1007 |
Instruction Following Codellama 70b Instruct leads
Codellama 70b Instruct: 51.9 (#293), Llama 2-7B: 50.8 (#298)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1024 | 1006 |
Long Context Not comparable
Codellama 70b Instruct: —, Llama 2-7B: 30.4 (#287)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | — | 999 |
Writing & Preference Codellama 70b Instruct leads
Codellama 70b Instruct: 33.4 (#277), Llama 2-7B: 28.0 (#298)
| Benchmark | Codellama 70b Instruct | Llama 2-7B |
|---|---|---|
| LMArena Text | 1057 | 1053 |
| LMArena Creative Writing | — | 1033 |
| LMArena Multi-Turn | — | 1029 |
Frequently asked questions
Is Codellama 70b Instruct better than Llama 2-7B?
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 29.1 on the Noometry Index.
Is Codellama 70b Instruct or Llama 2-7B better for coding?
Codellama 70b Instruct scores higher on coding benchmarks: 37.6 versus 29.2 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and Llama 2-7B share?
4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Llama 2-7B has 29.