Model comparison
Codellama 70b Instruct vs Llama 3-8B
Codellama 70b Instruct is the stronger model overall, scoring 33.3 to 25.3 on the Noometry Index.
Last verified . 7 shared benchmarks.
MetaSummary
- They share 7 benchmarks with published results for both. Codellama 70b Instruct scores higher in 2 categories and Llama 3-8B in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Llama 3-8B leads 58.7 to 52.1.
- The biggest single-benchmark swing is BigCodeBench Complete: 49.6% for Codellama 70b Instruct and 36.9% for Llama 3-8B.
Side by side
| Codellama 70b Instruct | Llama 3-8B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 33.3 | 25.3 |
| Released | — | 2024-04-18 |
| Weights | Open | Open |
| Context window | — | 8K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.14 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 7 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Codellama 70b Instruct leads
Codellama 70b Instruct: 37.0 (#189), Llama 3-8B: 30.5 (#285)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| BigCodeBench Instruct | 40.7% | 31.9% |
| BigCodeBench Complete | 49.6% | 36.9% |
| HumanEval+ | 65.9% | 56.7% |
| LMArena Coding | — | 1152 |
| MBPP+ | — | 54.8% |
Reasoning Codellama 70b Instruct leads
Codellama 70b Instruct: 19.2 (#241), Llama 3-8B: 13.8 (#321)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1133 |
| Chess Puzzles | — | 0% |
| DTBench | — | 43.9% |
| Adversarial NLI | — | 57.3% |
| Epoch Capabilities Index | — | 116.33 |
| ForecastBench | — | 58.6 |
| WinoGrande | — | 75.7% |
Math Not comparable
Codellama 70b Instruct: —, Llama 3-8B: 8.3 (#317)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 1.9% |
| LMArena Math | — | 1151 |
| MATH Level 5 | — | 6.1% |
Knowledge Not comparable
Codellama 70b Instruct: —, Llama 3-8B: 7.8 (#303)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| GPQA Diamond | — | 26.1% |
| LMArena Expert | — | 1113 |
| ARC (AI2) Challenge | — | 82.8% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 82.6% |
| TriviaQA | — | 67.7% |
Multilingual Llama 3-8B leads
Codellama 70b Instruct: 24.9 (#283), Llama 3-8B: 31.0 (#256)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Non-English | 992 | 1098 |
| LMArena Chinese | — | 1076 |
| LMArena French | — | 1158 |
| LMArena German | — | 1104 |
| LMArena Japanese | — | 967 |
| LMArena Korean | — | 1003 |
| LMArena Russian | — | 1109 |
| LMArena Spanish | — | 1172 |
Instruction Following Llama 3-8B leads
Codellama 70b Instruct: 52.1 (#288), Llama 3-8B: 58.7 (#255)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Instruction Following | 1024 | 1127 |
Long Context Not comparable
Codellama 70b Instruct: —, Llama 3-8B: 34.2 (#246)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Longer Query | — | 1128 |
Writing & Preference Llama 3-8B leads
Codellama 70b Instruct: 33.6 (#272), Llama 3-8B: 37.7 (#251)
| Benchmark | Codellama 70b Instruct | Llama 3-8B |
|---|---|---|
| LMArena Text | 1057 | 1166 |
| LMArena Creative Writing | — | 1150 |
| LMArena Multi-Turn | — | 1152 |
Frequently asked questions
Is Codellama 70b Instruct better than Llama 3-8B?
Codellama 70b Instruct is the stronger model overall, scoring 33.3 to 25.3 on the Noometry Index.
Is Codellama 70b Instruct or Llama 3-8B better for coding?
Codellama 70b Instruct scores higher on coding benchmarks: 37.0 versus 30.5 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and Llama 3-8B share?
7 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Llama 3-8B has 34.