# Llama 2-70B vs StarCoder 2 3B

> Llama 2-70B has enough public results to be ranked (#349); StarCoder 2 3B does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/llama-2-70b-vs-starcoder-2-3b
- Last updated: 2026-10-11
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and StarCoder 2 3B in 1 category; one gap is clear of the uncertainty.

## Snapshot

| | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 24.4 | 36.3 |
| Rank | 349 | — |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Llama 2-70B: 31.4 (#286)
- StarCoder 2 3B: 33.5

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| LMArena Coding | 1079 | — |
| BigCodeBench Complete | — | 21.4% |
| HumanEval+ | — | 27.4% |

## Reasoning

- Llama 2-70B: 14.4 (#325)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| Epoch Capabilities Index | 113.79 | 88.76 |
| WinoGrande | 80.2% | 57.1% |
| 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

- Llama 2-70B: 8.1 (#326)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| GSM8K | 69.6% | 21.6% |
| OTIS Mock AIME 2024-2025 | 0% | — |
| LMArena Math | 1091 | — |
| MATH Level 5 | 3.3% | — |

## Knowledge

- Llama 2-70B: 7.4 (#310)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| ARC (AI2) Challenge | 78.3% | 34.2% |
| MMLU | 69.9% | 36.6% |
| GPQA Diamond | 26.3% | — |
| LMArena Expert | 1039 | — |
| BoolQ | 88.6% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |

## Multilingual

- Llama 2-70B: 27.7 (#274)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| LMArena Non-English | 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: 54.9 (#278)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| LMArena Instruction Following | 1071 | — |

## Long Context

- Llama 2-70B: 32.3 (#270)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| LMArena Longer Query | 1062 | — |

## Writing & Preference

- Llama 2-70B: 32.3 (#279)
- StarCoder 2 3B: —

| Benchmark | Llama 2-70B | StarCoder 2 3B |
|---|---|---|
| LMArena Text | 1115 | — |
| LMArena Creative Writing | 1075 | — |
| LMArena Multi-Turn | 1088 | — |

## FAQ

### Is Llama 2-70B better than StarCoder 2 3B?

Llama 2-70B has enough public results to be ranked (#349); StarCoder 2 3B does not yet, so treat this comparison as directional.

### Is Llama 2-70B or StarCoder 2 3B better for coding?

StarCoder 2 3B scores higher on coding benchmarks: 33.5 versus 31.4 in the Noometry coding category.

### How many benchmarks do Llama 2-70B and StarCoder 2 3B share?

5 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and StarCoder 2 3B has 7.