# Llama 3.2 90B vs StarCoder 2 3B

> Llama 3.2 90B has enough public results to be ranked (#326); StarCoder 2 3B does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/llama-3-2-90b-vs-starcoder-2-3b
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both.

## Snapshot

| | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 27.1 | 35.8 |
| Rank | 326 | — |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Llama 3.2 90B: —
- StarCoder 2 3B: 32.9

| Benchmark | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| BigCodeBench Complete | — | 21.4% |
| HumanEval+ | — | 27.4% |

## Agentic & Tool Use

- Llama 3.2 90B: 29.7 (#73)
- StarCoder 2 3B: —

| Benchmark | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| BALROG | 27.3% | — |

## Reasoning

- Llama 3.2 90B: 20.9 (#207)
- StarCoder 2 3B: —

| Benchmark | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| Epoch Capabilities Index | 125.5 | 88.01 |
| EnigmaEval | 0.4% | — |
| WinoGrande | — | 57.1% |

## Math

- Llama 3.2 90B: 10.4 (#302)
- StarCoder 2 3B: —

| Benchmark | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.6% | — |
| MATH Level 5 | 39.4% | — |
| GSM8K | — | 21.6% |

## Knowledge

- Llama 3.2 90B: 21.7 (#269)
- StarCoder 2 3B: —

| Benchmark | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| MMLU | 80.3% | 36.6% |
| GPQA Diamond | 41% | — |
| ARC (AI2) Challenge | — | 34.2% |

## Multimodal

- Llama 3.2 90B: 24.9 (#119)
- StarCoder 2 3B: —

| Benchmark | Llama 3.2 90B | StarCoder 2 3B |
|---|---|---|
| LMArena Vision | 1000 | — |
| GeoBench | 52% | — |

## FAQ

### Is Llama 3.2 90B better than StarCoder 2 3B?

Llama 3.2 90B has enough public results to be ranked (#326); StarCoder 2 3B does not yet, so treat this comparison as directional.

### How many benchmarks do Llama 3.2 90B and StarCoder 2 3B share?

2 benchmarks have published results for both models. Llama 3.2 90B has 9 scored results on Noometry and StarCoder 2 3B has 7.