# GPT-6 Luna vs StarCoder 2 3B

> GPT-6 Luna has enough public results to be ranked (#65); StarCoder 2 3B does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/gpt-6-luna-vs-starcoder-2-3b
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in coding, where GPT-6 Luna leads 54.7 to 32.9.
- StarCoder 2 3B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| Provider | OpenAI | NVIDIA |
| Noometry Index | 45.1 | 35.8 |
| Rank | 65 | — |
| Context | 1.05M | — |
| Input $/M | $0.10 | — |
| Output $/M | $0.50 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Luna: 54.7 (#24)
- StarCoder 2 3B: 32.9

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1583 | — |
| LMArena Coding | 1448 | — |
| BigCodeBench Complete | — | 21.4% |
| HumanEval+ | — | 27.4% |

## Reasoning

- GPT-6 Luna: 31.7 (#68)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| NYT Connections (extended) | 68.7% | — |
| LMArena Hard Prompts | 1415 | — |
| Epoch Capabilities Index | — | 88.01 |
| WinoGrande | — | 57.1% |

## Math

- GPT-6 Luna: 35.6 (#140)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| LMArena Math | 1396 | — |
| GSM8K | — | 21.6% |

## Knowledge

- GPT-6 Luna: 40.1 (#110)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| LMArena Expert | 1458 | — |
| ARC (AI2) Challenge | — | 34.2% |
| MMLU | — | 36.6% |

## Multimodal

- GPT-6 Luna: 42.0 (#27)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| Furniture Assembly | 44.2% | — |

## Multilingual

- GPT-6 Luna: 51.0 (#111)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| LMArena Non-English | 1390 | — |
| LMArena Chinese | 1425 | — |
| LMArena Russian | 1392 | — |
| LMArena Spanish | 1416 | — |

## Instruction Following

- GPT-6 Luna: 74.7 (#85)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| LMArena Instruction Following | 1414 | — |

## Long Context

- GPT-6 Luna: 43.3 (#95)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| LMArena Longer Query | 1417 | — |

## Writing & Preference

- GPT-6 Luna: 58.4 (#115)
- StarCoder 2 3B: —

| Benchmark | GPT-6 Luna | StarCoder 2 3B |
|---|---|---|
| LMArena Text | 1392 | — |
| LMArena Creative Writing | 1357 | — |
| LMArena Multi-Turn | 1400 | — |

## FAQ

### Is GPT-6 Luna better than StarCoder 2 3B?

GPT-6 Luna has enough public results to be ranked (#65); StarCoder 2 3B does not yet, so treat this comparison as directional.

### Is GPT-6 Luna or StarCoder 2 3B better for coding?

GPT-6 Luna scores higher on coding benchmarks: 54.7 versus 32.9 in the Noometry coding category.

### How many benchmarks do GPT-6 Luna and StarCoder 2 3B share?

0 benchmarks have published results for both models. GPT-6 Luna has 18 scored results on Noometry and StarCoder 2 3B has 7.