Model comparison
Llama 2-70B vs StarCoder 2 15B
Llama 2-70B has enough public results to be ranked (#343); StarCoder 2 15B does not yet, so treat this comparison as directional.
Last verified . 5 shared benchmarks.
Meta
NVIDIASummary
- They share 5 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and StarCoder 2 15B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where StarCoder 2 15B leads 36.0 to 30.8.
Side by side
| Llama 2-70B | StarCoder 2 15B | |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 24.1 | 37.5 |
| Released | 2023-07-18 | 2024-02-20 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 15B leads
Llama 2-70B: 30.8 (#281), StarCoder 2 15B: 36.0
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| LMArena Coding | 1079 | — |
| BigCodeBench Complete | — | 38.4% |
| HumanEval+ | — | 37.8% |
Reasoning Not comparable
Llama 2-70B: 13.7 (#324), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| Epoch Capabilities Index | 113.59 | 104.61 |
| WinoGrande | 80.2% | 64.3% |
| LMArena Hard Prompts | 1073 | — |
| DTBench | 41.6% | — |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
Math Not comparable
Llama 2-70B: 7.5 (#320), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| GSM8K | 69.6% | 57.7% |
| OTIS Mock AIME 2024-2025 | 0% | — |
| LMArena Math | 1091 | — |
| MATH Level 5 | 3.3% | — |
Knowledge Not comparable
Llama 2-70B: 7.4 (#305), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| ARC (AI2) Challenge | 78.3% | 47.2% |
| MMLU | 69.9% | 64.1% |
| GPQA Diamond | 26.3% | — |
| LMArena Expert | 1039 | — |
| BoolQ | 88.6% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Not comparable
Llama 2-70B: 27.9 (#269), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| LMArena Non-English | 1045 | — |
| LMArena Chinese | 994 | — |
| LMArena French | 1089 | — |
| LMArena German | 1041 | — |
| LMArena Japanese | 927 | — |
| LMArena Korean | 964 | — |
| LMArena Russian | 1083 | — |
| LMArena Spanish | 1142 | — |
Instruction Following Not comparable
Llama 2-70B: 55.1 (#273), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| LMArena Instruction Following | 1071 | — |
Long Context Not comparable
Llama 2-70B: 32.2 (#265), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| LMArena Longer Query | 1062 | — |
Writing & Preference Not comparable
Llama 2-70B: 32.5 (#274), StarCoder 2 15B: —
| Benchmark | Llama 2-70B | StarCoder 2 15B |
|---|---|---|
| LMArena Text | 1115 | — |
| LMArena Creative Writing | 1075 | — |
| LMArena Multi-Turn | 1088 | — |
Frequently asked questions
Is Llama 2-70B better than StarCoder 2 15B?
Llama 2-70B has enough public results to be ranked (#343); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is Llama 2-70B or StarCoder 2 15B better for coding?
StarCoder 2 15B scores higher on coding benchmarks: 36.0 versus 30.8 in the Noometry coding category.
How many benchmarks do Llama 2-70B and StarCoder 2 15B share?
5 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and StarCoder 2 15B has 7.