# Qwen3-Coder 480B-A35B Instruct vs Step 3

> Step 3 is the stronger model overall, scoring 40.5 to 38.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen3-coder-480b-a35b-instruct-vs-step-3
- Last updated: 2026-10-11
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Qwen3-Coder 480B-A35B Instruct scores higher in 5 categories and Step 3 in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 3 leads 40.1 to 35.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 49.5% for Qwen3-Coder 480B-A35B Instruct and 62.3% for Step 3.

## Snapshot

| | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 38.1 | 40.5 |
| Rank | 190 | 149 |
| Context | 262K | — |
| Input $/M | $1.50 | — |
| Output $/M | $7.50 | — |
| Weights | Open | Open |

## Coding

- Qwen3-Coder 480B-A35B Instruct: 35.5 (#223)
- Step 3: 40.1 (#147)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Coding | 1412 | 1367 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1275 | — |
| GSO | 4.9% | — |
| WeirdML | 41.2% | — |
| ALE-Bench | 461.45 | — |
| AlgoTune | 1.44 | — |

## Agentic & Tool Use

- Qwen3-Coder 480B-A35B Instruct: 23.9 (#123)
- Step 3: —

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| Terminal-Bench | 27.2% | — |

## Reasoning

- Qwen3-Coder 480B-A35B Instruct: 25.5 (#149)
- Step 3: 28.4 (#105)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 49.5% | 62.3% |
| LMArena Hard Prompts | 1372 | 1355 |

## Math

- Qwen3-Coder 480B-A35B Instruct: 37.6 (#150)
- Step 3: 37.6 (#148)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Math | 1365 | 1366 |

## Knowledge

- Qwen3-Coder 480B-A35B Instruct: 37.0 (#162)
- Step 3: 36.8 (#164)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Expert | 1338 | 1333 |

## Multimodal

- Qwen3-Coder 480B-A35B Instruct: —
- Step 3: 35.5 (#86)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |

## Multilingual

- Qwen3-Coder 480B-A35B Instruct: 47.7 (#148)
- Step 3: 46.3 (#159)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Non-English | 1346 | 1327 |
| LMArena Chinese | 1357 | 1397 |
| LMArena German | 1325 | 1371 |
| LMArena Korean | 1305 | 1269 |
| LMArena Russian | 1366 | 1331 |
| LMArena Spanish | 1360 | 1371 |
| LMArena French | 1398 | — |
| LMArena Japanese | 1310 | — |

## Instruction Following

- Qwen3-Coder 480B-A35B Instruct: 71.6 (#147)
- Step 3: 70.4 (#164)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1355 | 1332 |

## Long Context

- Qwen3-Coder 480B-A35B Instruct: 42.0 (#131)
- Step 3: 40.3 (#157)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Longer Query | 1378 | 1326 |

## Writing & Preference

- Qwen3-Coder 480B-A35B Instruct: 55.3 (#147)
- Step 3: 54.3 (#151)

| Benchmark | Qwen3-Coder 480B-A35B Instruct | Step 3 |
|---|---|---|
| LMArena Text | 1357 | 1350 |
| LMArena Creative Writing | 1333 | 1321 |
| LMArena Multi-Turn | 1365 | 1341 |

## FAQ

### Is Qwen3-Coder 480B-A35B Instruct better than Step 3?

Step 3 is the stronger model overall, scoring 40.5 to 38.1 on the Noometry Index.

### Is Qwen3-Coder 480B-A35B Instruct or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 40.1 versus 35.5 in the Noometry coding category.

### How many benchmarks do Qwen3-Coder 480B-A35B Instruct and Step 3 share?

16 benchmarks have published results for both models. Qwen3-Coder 480B-A35B Instruct has 25 scored results on Noometry and Step 3 has 17.