# GPT-6 Luna vs Qwen3-Coder 480B-A35B Instruct

> GPT-6 Luna is the stronger model overall, scoring 53.3 to 38.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-6-luna-vs-qwen3-coder-480b-a35b-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. GPT-6 Luna scores higher in 9 categories and Qwen3-Coder 480B-A35B Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 37.6.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $1.50 / $7.50 for Qwen3-Coder 480B-A35B Instruct.
- GPT-6 Luna accepts more context: 1.05M tokens versus 262K.
- Qwen3-Coder 480B-A35B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 53.3 | 38.1 |
| Rank | 36 | 190 |
| Context | 1.05M | 262K |
| Input $/M | $0.10 | $1.50 |
| Output $/M | $0.50 | $7.50 |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Luna: 55.5 (#25)
- Qwen3-Coder 480B-A35B Instruct: 35.5 (#223)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena WebDev | 1581 | 1275 |
| LMArena Coding | 1439 | 1412 |
| ALE-Bench | 1,577 | 461.45 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| SWE-bench Verified (bash only) | — | 55.4% |
| SciCode | 54.6% | — |
| GSO | — | 4.9% |
| WeirdML | — | 41.2% |
| AlgoTune | — | 1.44 |

## Agentic & Tool Use

- GPT-6 Luna: 33.3 (#54)
- Qwen3-Coder 480B-A35B Instruct: 23.9 (#123)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| APEX-Agents | 44.3% | — |
| GDP.pdf | 23% | — |

## Reasoning

- GPT-6 Luna: 48.2 (#41)
- Qwen3-Coder 480B-A35B Instruct: 25.5 (#149)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1411 | 1372 |
| ARC-AGI-2 | 59.3% | — |
| Kagi LLM Benchmark | — | 49.5% |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| CritPt | 19.4% | — |
| Chess Puzzles | 31% | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 90.1% | — |
| LMCA | 44.5% | — |
| Epoch Capabilities Index | 156.28 | — |

## Math

- GPT-6 Luna: 76.1 (#15)
- Qwen3-Coder 480B-A35B Instruct: 37.6 (#150)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1416 | 1365 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 64% | — |

## Knowledge

- GPT-6 Luna: 57.0 (#41)
- Qwen3-Coder 480B-A35B Instruct: 37.0 (#162)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1444 | 1338 |
| GPQA Diamond | 90.5% | — |
| SimpleQA Verified | 41.4% | — |

## Multimodal

- GPT-6 Luna: 42.4 (#30)
- Qwen3-Coder 480B-A35B Instruct: —

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |

## Multilingual

- GPT-6 Luna: 50.5 (#117)
- Qwen3-Coder 480B-A35B Instruct: 47.7 (#148)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1386 | 1346 |
| LMArena Chinese | 1433 | 1357 |
| LMArena French | 1420 | 1398 |
| LMArena German | 1369 | 1325 |
| LMArena Japanese | 1369 | 1310 |
| LMArena Korean | 1360 | 1305 |
| LMArena Russian | 1394 | 1366 |
| LMArena Spanish | 1393 | 1360 |

## Instruction Following

- GPT-6 Luna: 74.3 (#99)
- Qwen3-Coder 480B-A35B Instruct: 71.6 (#147)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1409 | 1355 |

## Long Context

- GPT-6 Luna: 43.0 (#111)
- Qwen3-Coder 480B-A35B Instruct: 42.0 (#131)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1409 | 1378 |

## Writing & Preference

- GPT-6 Luna: 58.3 (#119)
- Qwen3-Coder 480B-A35B Instruct: 55.3 (#147)

| Benchmark | GPT-6 Luna | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1391 | 1357 |
| LMArena Creative Writing | 1363 | 1333 |
| LMArena Multi-Turn | 1396 | 1365 |

## FAQ

### Is GPT-6 Luna better than Qwen3-Coder 480B-A35B Instruct?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 38.1 on the Noometry Index.

### Which is cheaper, GPT-6 Luna or Qwen3-Coder 480B-A35B Instruct?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.

### Is GPT-6 Luna or Qwen3-Coder 480B-A35B Instruct better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 35.5 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 262K.

### How many benchmarks do GPT-6 Luna and Qwen3-Coder 480B-A35B Instruct share?

19 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.