# GPT-4.1 nano vs Qwen2.5-Coder-32B

> Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 27.9 on the Noometry Index. GPT-4.1 nano costs 4.3× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4-1-nano-vs-qwen2-5-coder-32b
- Last updated: 2026-10-12
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. GPT-4.1 nano scores higher in 3 categories and Qwen2.5-Coder-32B in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Qwen2.5-Coder-32B leads 38.0 to 23.7.
- The biggest single-benchmark swing is Aider Polyglot: 8.9% for GPT-4.1 nano and 16.4% for Qwen2.5-Coder-32B.
- GPT-4.1 nano is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- GPT-4.1 nano accepts more context: 1.05M tokens versus 33K.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 27.9 | 33.4 |
| Rank | 327 | 245 |
| Context | 1.05M | 33K |
| Input $/M | $0.10 | $0.66 |
| Output $/M | $0.40 | $1 |
| Weights | Proprietary | Open |

## Coding

- GPT-4.1 nano: 24.1 (#330)
- Qwen2.5-Coder-32B: 22.6 (#333)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| Aider Polyglot | 8.9% | 16.4% |
| LMArena Coding | 1306 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| SciCode | 25.9% | — |
| WeirdML | 19% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |

## Agentic & Tool Use

- GPT-4.1 nano: 26.5 (#104)
- Qwen2.5-Coder-32B: —

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 33% | — |

## Reasoning

- GPT-4.1 nano: 8.5 (#349)
- Qwen2.5-Coder-32B: 21.2 (#225)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1286 | 1251 |
| Epoch Capabilities Index | 129.62 | 119.49 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 33.3% | — |
| ARC-AGI-1 | 0% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 52.5% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 5.5% | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |

## Math

- GPT-4.1 nano: 26.9 (#252)
- Qwen2.5-Coder-32B: 33.3 (#204)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1274 | 1251 |
| OTIS Mock AIME 2024-2025 | 28.9% | — |
| Omni-MATH | 36.7% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 70% | — |
| FrontierMath (Feb 2025 set) | 1% | — |
| GSM8K | — | 93% |

## Knowledge

- GPT-4.1 nano: 21.8 (#273)
- Qwen2.5-Coder-32B: 33.4 (#203)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1272 | 1221 |
| GPQA Diamond | 48.9% | — |
| SimpleQA Verified | 6% | — |
| MMLU-Pro | 55% | — |
| GPQA (HELM) | 50.7% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |

## Multimodal

- GPT-4.1 nano: 29.2 (#113)
- Qwen2.5-Coder-32B: —

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1063 | — |

## Multilingual

- GPT-4.1 nano: 41.6 (#205)
- Qwen2.5-Coder-32B: 37.8 (#235)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1260 | 1205 |
| LMArena Chinese | 1270 | 1222 |
| LMArena Russian | 1261 | 1228 |
| LMArena German | 1288 | — |
| LMArena Japanese | 1198 | — |

## Instruction Following

- GPT-4.1 nano: 67.8 (#193)
- Qwen2.5-Coder-32B: 61.4 (#245)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1267 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
| IFEval | 84.3% | — |

## Long Context

- GPT-4.1 nano: 23.7 (#296)
- Qwen2.5-Coder-32B: 38.0 (#208)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1283 | 1251 |
| Fiction.LiveBench | 25% | — |

## Writing & Preference

- GPT-4.1 nano: 40.5 (#243)
- Qwen2.5-Coder-32B: 41.6 (#240)

| Benchmark | GPT-4.1 nano | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1285 | 1230 |
| LMArena Creative Writing | 1260 | 1174 |
| LMArena Multi-Turn | 1277 | 1222 |
| EQ-Bench Creative Writing | 946 | — |
| WildBench | 81.2% | — |
| LiveBench Language | — | 23.3% |

## FAQ

### Is GPT-4.1 nano better than Qwen2.5-Coder-32B?

Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 27.9 on the Noometry Index. GPT-4.1 nano costs 4.3× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.

### Which is cheaper, GPT-4.1 nano or Qwen2.5-Coder-32B?

GPT-4.1 nano is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.

### Is GPT-4.1 nano or Qwen2.5-Coder-32B better for coding?

GPT-4.1 nano scores higher on coding benchmarks: 24.1 versus 22.6 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 nano does, with 1.05M tokens against 33K.

### How many benchmarks do GPT-4.1 nano and Qwen2.5-Coder-32B share?

14 benchmarks have published results for both models. GPT-4.1 nano has 38 scored results on Noometry and Qwen2.5-Coder-32B has 31.