# Qwen2.5-Coder-32B vs Qwen3-Coder 480B-A35B Instruct

> Qwen3-Coder 480B-A35B Instruct is the stronger model overall, scoring 37.7 to 32.9 on the Noometry Index. Qwen2.5-Coder-32B costs 4.0× less per token, which makes it the better buy when Qwen3-Coder 480B-A35B Instruct's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/qwen2-5-coder-32b-vs-qwen3-coder-480b-a35b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 13

## Summary

- They share 13 benchmarks with published results for both. Qwen2.5-Coder-32B scores higher in 0 categories and Qwen3-Coder 480B-A35B Instruct in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen3-Coder 480B-A35B Instruct leads 36.2 to 22.4.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 9% for Qwen2.5-Coder-32B and 55.4% for Qwen3-Coder 480B-A35B Instruct.
- Qwen2.5-Coder-32B is cheaper at $0.66 / $1 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.

## Snapshot

| | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 32.9 | 37.7 |
| Rank | 239 | 186 |
| Context | 33K | 262K |
| Input $/M | $0.66 | $1.50 |
| Output $/M | $1 | $7.50 |
| Weights | Open | Open |

## Coding

- Qwen2.5-Coder-32B: 22.4 (#328)
- Qwen3-Coder 480B-A35B Instruct: 36.2 (#199)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| SWE-bench Verified (bash only) | 9% | 55.4% |
| LMArena Coding | 1276 | 1413 |
| Aider Polyglot | 16.4% | — |
| LMArena WebDev | — | 1275 |
| GSO | — | 4.9% |
| WeirdML | — | 41.2% |
| BigCodeBench Instruct | 49% | — |
| LiveBench Coding | 56.9% | — |
| BigCodeBench Complete | 58% | — |
| ALE-Bench | — | 461.45 |
| AlgoTune | — | 1.44 |
| HumanEval+ | 87.2% | — |
| MBPP+ | 77% | — |

## Agentic & Tool Use

- Qwen2.5-Coder-32B: —
- Qwen3-Coder 480B-A35B Instruct: 23.7 (#116)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |

## Reasoning

- Qwen2.5-Coder-32B: 20.4 (#220)
- Qwen3-Coder 480B-A35B Instruct: 24.7 (#141)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1372 |
| Kagi LLM Benchmark | — | 49.5% |
| LiveBench Reasoning | 42.1% | — |
| LiveBench Data Analysis | 49.9% | — |
| Epoch Capabilities Index | 119.41 | — |
| HellaSwag | 83% | — |
| LiveBench | 46.2% | — |
| WinoGrande | 80.8% | — |

## Math

- Qwen2.5-Coder-32B: 30.8 (#203)
- Qwen3-Coder 480B-A35B Instruct: 34.9 (#151)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1251 | 1364 |
| LiveBench Math | 46.6% | — |
| GSM8K | 93% | — |

## Knowledge

- Qwen2.5-Coder-32B: 32.9 (#200)
- Qwen3-Coder 480B-A35B Instruct: 36.4 (#159)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1221 | 1337 |
| ARC (AI2) Challenge | 70.5% | — |
| MMLU | 79.1% | — |

## Multilingual

- Qwen2.5-Coder-32B: 38.0 (#230)
- Qwen3-Coder 480B-A35B Instruct: 47.8 (#143)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1205 | 1345 |
| LMArena Chinese | 1222 | 1356 |
| LMArena Russian | 1228 | 1367 |
| LMArena French | — | 1394 |
| LMArena German | — | 1324 |
| LMArena Japanese | — | 1309 |
| LMArena Korean | — | 1307 |
| LMArena Spanish | — | 1360 |

## Instruction Following

- Qwen2.5-Coder-32B: 61.6 (#240)
- Qwen3-Coder 480B-A35B Instruct: 71.8 (#142)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1223 | 1356 |
| LiveBench Instruction Following | 58.7% | — |

## Long Context

- Qwen2.5-Coder-32B: 37.9 (#204)
- Qwen3-Coder 480B-A35B Instruct: 42.0 (#126)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1251 | 1379 |

## Writing & Preference

- Qwen2.5-Coder-32B: 41.8 (#235)
- Qwen3-Coder 480B-A35B Instruct: 55.5 (#142)

| Benchmark | Qwen2.5-Coder-32B | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1230 | 1356 |
| LMArena Creative Writing | 1174 | 1333 |
| LMArena Multi-Turn | 1222 | 1365 |
| LiveBench Language | 23.3% | — |

## FAQ

### Is Qwen2.5-Coder-32B better than Qwen3-Coder 480B-A35B Instruct?

Qwen3-Coder 480B-A35B Instruct is the stronger model overall, scoring 37.7 to 32.9 on the Noometry Index. Qwen2.5-Coder-32B costs 4.0× less per token, which makes it the better buy when Qwen3-Coder 480B-A35B Instruct's lead doesn't matter for your workload.

### Which is cheaper, Qwen2.5-Coder-32B or Qwen3-Coder 480B-A35B Instruct?

Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.

### Is Qwen2.5-Coder-32B or Qwen3-Coder 480B-A35B Instruct better for coding?

Qwen3-Coder 480B-A35B Instruct scores higher on coding benchmarks: 36.2 versus 22.4 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 Qwen2.5-Coder-32B and Qwen3-Coder 480B-A35B Instruct share?

13 benchmarks have published results for both models. Qwen2.5-Coder-32B has 31 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.