# Qwen Max vs Qwen3-Coder 480B-A35B Instruct

> Qwen3-Coder 480B-A35B Instruct is the stronger model overall, scoring 37.7 to 34.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen-max-vs-qwen3-coder-480b-a35b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Qwen Max scores higher in 0 categories and Qwen3-Coder 480B-A35B Instruct in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3-Coder 480B-A35B Instruct leads 34.9 to 21.0.
- Qwen Max is cheaper at $1.60 / $6.40 per million input/output tokens, against $1.50 / $7.50 for Qwen3-Coder 480B-A35B Instruct.
- Qwen3-Coder 480B-A35B Instruct accepts more context: 262K tokens versus 33K.
- Qwen3-Coder 480B-A35B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 34.4 | 37.7 |
| Rank | 226 | 186 |
| Context | 33K | 262K |
| Input $/M | $1.60 | $1.50 |
| Output $/M | $6.40 | $7.50 |
| Weights | Proprietary | Open |

## Coding

- Qwen Max: 30.3 (#288)
- Qwen3-Coder 480B-A35B Instruct: 36.2 (#199)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Coding | 1288 | 1413 |
| SWE-bench Verified (bash only) | — | 55.4% |
| Aider Polyglot | 21.8% | — |
| LMArena WebDev | — | 1275 |
| GSO | — | 4.9% |
| WeirdML | — | 41.2% |
| ALE-Bench | — | 461.45 |
| AlgoTune | — | 1.44 |

## Agentic & Tool Use

- Qwen Max: —
- Qwen3-Coder 480B-A35B Instruct: 23.7 (#116)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |

## Reasoning

- Qwen Max: 24.1 (#154)
- Qwen3-Coder 480B-A35B Instruct: 24.7 (#141)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1270 | 1372 |
| Kagi LLM Benchmark | — | 49.5% |

## Math

- Qwen Max: 21.0 (#272)
- Qwen3-Coder 480B-A35B Instruct: 34.9 (#151)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1275 | 1364 |
| OTIS Mock AIME 2024-2025 | 16.1% | — |
| MATH Level 5 | 67.2% | — |
| FrontierMath (Feb 2025 set) | 1% | — |

## Knowledge

- Qwen Max: 30.3 (#222)
- Qwen3-Coder 480B-A35B Instruct: 36.4 (#159)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1248 | 1337 |
| GPQA Diamond | 56.1% | — |

## Multilingual

- Qwen Max: 42.0 (#197)
- Qwen3-Coder 480B-A35B Instruct: 47.8 (#143)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1263 | 1345 |
| LMArena Chinese | 1254 | 1356 |
| LMArena French | 1329 | 1394 |
| LMArena German | 1254 | 1324 |
| LMArena Japanese | 1205 | 1309 |
| LMArena Korean | 1141 | 1307 |
| LMArena Russian | 1274 | 1367 |
| LMArena Spanish | 1290 | 1360 |

## Instruction Following

- Qwen Max: 66.7 (#203)
- Qwen3-Coder 480B-A35B Instruct: 71.8 (#142)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1262 | 1356 |

## Long Context

- Qwen Max: 39.4 (#175)
- Qwen3-Coder 480B-A35B Instruct: 42.0 (#126)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1289 | 1379 |
| Fiction.LiveBench | 66.7% | — |

## Writing & Preference

- Qwen Max: 48.0 (#200)
- Qwen3-Coder 480B-A35B Instruct: 55.5 (#142)

| Benchmark | Qwen Max | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1282 | 1356 |
| LMArena Creative Writing | 1248 | 1333 |
| LMArena Multi-Turn | 1277 | 1365 |

## FAQ

### Is Qwen Max better than Qwen3-Coder 480B-A35B Instruct?

Qwen3-Coder 480B-A35B Instruct is the stronger model overall, scoring 37.7 to 34.4 on the Noometry Index.

### Which is cheaper, Qwen Max or Qwen3-Coder 480B-A35B Instruct?

Qwen Max is cheaper. It lists at $1.60 per million input tokens and $6.40 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.

### Is Qwen Max or Qwen3-Coder 480B-A35B Instruct better for coding?

Qwen3-Coder 480B-A35B Instruct scores higher on coding benchmarks: 36.2 versus 30.3 in the Noometry coding category.

### Which has the bigger context window?

Qwen3-Coder 480B-A35B Instruct does, with 262K tokens against 33K.

### How many benchmarks do Qwen Max and Qwen3-Coder 480B-A35B Instruct share?

17 benchmarks have published results for both models. Qwen Max has 23 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.