Model comparison
Codellama 70b Instruct vs Llama 2-70B
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 24.4 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 3 categories and Llama 2-70B in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where Codellama 70b Instruct leads 37.6 to 31.4.
Side by side
| Codellama 70b Instruct | Llama 2-70B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 33.7 | 24.4 |
| Released | — | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 7 | 35 |
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 2-70B: 31.4 (#286)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1079 |
| BigCodeBench Complete | 49.6% | — |
| HumanEval+ | 65.9% | — |
Reasoning Codellama 70b Instruct leads
Codellama 70b Instruct: 20.1 (#242), Llama 2-70B: 14.4 (#325)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1073 |
| DTBench | — | 41.6% |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| Epoch Capabilities Index | — | 113.79 |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math Not comparable
Codellama 70b Instruct: —, Llama 2-70B: 8.1 (#326)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 0% |
| LMArena Math | — | 1091 |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge Not comparable
Codellama 70b Instruct: —, Llama 2-70B: 7.4 (#310)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| GPQA Diamond | — | 26.3% |
| LMArena Expert | — | 1039 |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual Llama 2-70B leads
Codellama 70b Instruct: 24.8 (#288), Llama 2-70B: 27.7 (#274)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 992 | 1045 |
| LMArena Chinese | — | 995 |
| LMArena French | — | 1090 |
| LMArena German | — | 1041 |
| LMArena Japanese | — | 927 |
| LMArena Korean | — | 964 |
| LMArena Russian | — | 1083 |
| LMArena Spanish | — | 1143 |
Instruction Following Llama 2-70B leads
Codellama 70b Instruct: 51.9 (#293), Llama 2-70B: 54.9 (#278)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1024 | 1071 |
Long Context Not comparable
Codellama 70b Instruct: —, Llama 2-70B: 32.3 (#270)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | — | 1062 |
Writing & Preference Codellama 70b Instruct leads
Codellama 70b Instruct: 33.4 (#277), Llama 2-70B: 32.3 (#279)
| Benchmark | Codellama 70b Instruct | Llama 2-70B |
|---|---|---|
| LMArena Text | 1057 | 1115 |
| LMArena Creative Writing | — | 1075 |
| LMArena Multi-Turn | — | 1088 |
Frequently asked questions
Is Codellama 70b Instruct better than Llama 2-70B?
Codellama 70b Instruct is the stronger model overall, scoring 33.7 to 24.4 on the Noometry Index.
Is Codellama 70b Instruct or Llama 2-70B better for coding?
Codellama 70b Instruct scores higher on coding benchmarks: 37.6 versus 31.4 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and Llama 2-70B share?
4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Llama 2-70B has 35.