Model comparison
Codellama 70b Instruct vs Llama 3.1-70B
Codellama 70b Instruct is the stronger model overall, scoring 33.3 to 29.4 on the Noometry Index.
Last verified . 6 shared benchmarks.
MetaSummary
- They share 6 benchmarks with published results for both. Codellama 70b Instruct scores higher in 1 category and Llama 3.1-70B in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Llama 3.1-70B leads 38.9 to 24.9.
- The biggest single-benchmark swing is BigCodeBench Instruct: 40.7% for Codellama 70b Instruct and 46.1% for Llama 3.1-70B.
Side by side
| Codellama 70b Instruct | Llama 3.1-70B | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 33.3 | 29.4 |
| Released | — | 2024-07-23 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.40 |
| Output $ / M tokens | — | $0.40 |
| 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: 37.0 (#189), Llama 3.1-70B: 29.8 (#292)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| BigCodeBench Instruct | 40.7% | 46.1% |
| BigCodeBench Complete | 49.6% | 54.8% |
| WeirdML | — | 9% |
| LMArena Coding | — | 1260 |
| HumanEval+ | 65.9% | — |
Agentic & Tool Use Not comparable
Codellama 70b Instruct: —, Llama 3.1-70B: 24.9 (#104)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |
Reasoning Llama 3.1-70B leads
Codellama 70b Instruct: 19.2 (#241), Llama 3.1-70B: 20.8 (#215)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1241 |
| DTBench | — | 60% |
| LMCA | — | 14.8% |
| Epoch Capabilities Index | — | 125.91 |
Math Not comparable
Codellama 70b Instruct: —, Llama 3.1-70B: 12.7 (#298)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 3.6% |
| Omni-MATH | — | 21% |
| LMArena Math | — | 1252 |
| MATH Level 5 | — | 36.7% |
Knowledge Not comparable
Codellama 70b Instruct: —, Llama 3.1-70B: 24.1 (#264)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| GPQA Diamond | — | 44.2% |
| MMLU-Pro | — | 65.3% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1209 |
| MMLU | — | 80.1% |
Multilingual Llama 3.1-70B leads
Codellama 70b Instruct: 24.9 (#283), Llama 3.1-70B: 38.9 (#220)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 992 | 1219 |
| LMArena Chinese | — | 1215 |
| LMArena French | — | 1260 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1132 |
| LMArena Korean | — | 1140 |
| LMArena Russian | — | 1234 |
| LMArena Spanish | — | 1253 |
Instruction Following Llama 3.1-70B leads
Codellama 70b Instruct: 52.1 (#288), Llama 3.1-70B: 65.5 (#218)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1024 | 1231 |
| IFEval | — | 82.1% |
Long Context Not comparable
Codellama 70b Instruct: —, Llama 3.1-70B: 37.6 (#208)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | — | 1241 |
Writing & Preference Llama 3.1-70B leads
Codellama 70b Instruct: 33.6 (#272), Llama 3.1-70B: 35.6 (#262)
| Benchmark | Codellama 70b Instruct | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1057 | 1261 |
| LMArena Creative Writing | — | 1232 |
| EQ-Bench Creative Writing | — | 784 |
| WildBench | — | 75.8% |
| LMArena Multi-Turn | — | 1256 |
Frequently asked questions
Is Codellama 70b Instruct better than Llama 3.1-70B?
Codellama 70b Instruct is the stronger model overall, scoring 33.3 to 29.4 on the Noometry Index.
Is Codellama 70b Instruct or Llama 3.1-70B better for coding?
Codellama 70b Instruct scores higher on coding benchmarks: 37.0 versus 29.8 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and Llama 3.1-70B share?
6 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Llama 3.1-70B has 35.