# o4-mini vs Qwen3-Coder 480B-A35B Instruct

> o4-mini is the stronger model overall, scoring 41.6 to 37.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/o4-mini-vs-qwen3-coder-480b-a35b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. o4-mini scores higher in 6 categories and Qwen3-Coder 480B-A35B Instruct in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where o4-mini leads 32.3 to 23.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 67.6% for o4-mini and 49.5% for Qwen3-Coder 480B-A35B Instruct.
- o4-mini is cheaper at $1.10 / $4.40 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 200K.
- Qwen3-Coder 480B-A35B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.6 | 37.7 |
| Rank | 119 | 186 |
| Context | 200K | 262K |
| Input $/M | $1.10 | $1.50 |
| Output $/M | $4.40 | $7.50 |
| Weights | Proprietary | Open |

## Coding

- o4-mini: 41.6 (#110)
- Qwen3-Coder 480B-A35B Instruct: 36.2 (#199)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| SWE-bench Verified (bash only) | 45% | 55.4% |
| GSO | 3.6% | 4.9% |
| WeirdML | 52.6% | 41.2% |
| LMArena Coding | 1369 | 1413 |
| ALE-Bench | 826.17 | 461.45 |
| AlgoTune | 1.72 | 1.44 |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1275 |
| CadEval | 62% | — |

## Agentic & Tool Use

- o4-mini: 32.3 (#57)
- Qwen3-Coder 480B-A35B Instruct: 23.7 (#116)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| Berkeley Function Calling Leaderboard | 53.2% | — |
| GDPval | 25.3% | — |
| METR Time Horizons | 63.9% | — |

## Reasoning

- o4-mini: 24.6 (#143)
- Qwen3-Coder 480B-A35B Instruct: 24.7 (#141)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 67.6% | 49.5% |
| LMArena Hard Prompts | 1351 | 1372 |
| ARC-AGI-2 | 6.1% | — |
| SimpleBench | 38.7% | — |
| ARC-AGI-1 | 58.7% | — |
| CritPt | 0.6% | — |
| Chess Puzzles | 26% | — |
| EnigmaEval | 9.2% | — |
| Mystery Game Puzzles | 5% | — |
| DTBench | 77.6% | — |
| LMCA | 26.5% | — |
| Epoch Capabilities Index | 145.63 | — |
| ForecastBench | 61.8 | — |

## Math

- o4-mini: 40.2 (#79)
- Qwen3-Coder 480B-A35B Instruct: 34.9 (#151)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1390 | 1364 |
| FrontierMath (Tiers 1-3) | 36.1% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 81.7% | — |
| Omni-MATH | 72% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 24.8% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |

## Knowledge

- o4-mini: 43.5 (#83)
- Qwen3-Coder 480B-A35B Instruct: 36.4 (#159)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1344 | 1337 |
| GPQA Diamond | 79.6% | — |
| Humanity's Last Exam | 18.1% | — |
| SimpleQA Verified | 19.6% | — |
| MMLU-Pro | 82% | — |
| Confabulations | 15.8% | — |
| Vectara Hallucination Rate | 18.6% | — |
| GPQA (HELM) | 73.5% | — |

## Multimodal

- o4-mini: 39.5 (#46)
- Qwen3-Coder 480B-A35B Instruct: —

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Vision | 1194 | — |
| GeoBench | 64% | — |
| VPCT | 57.5% | — |

## Multilingual

- o4-mini: 47.2 (#149)
- Qwen3-Coder 480B-A35B Instruct: 47.8 (#143)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1336 | 1345 |
| LMArena Chinese | 1355 | 1356 |
| LMArena French | 1362 | 1394 |
| LMArena German | 1336 | 1324 |
| LMArena Japanese | 1308 | 1309 |
| LMArena Korean | 1313 | 1307 |
| LMArena Russian | 1334 | 1367 |
| LMArena Spanish | 1346 | 1360 |

## Instruction Following

- o4-mini: 75.4 (#63)
- Qwen3-Coder 480B-A35B Instruct: 71.8 (#142)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1321 | 1356 |
| IFEval | 92.8% | — |

## Long Context

- o4-mini: 45.5 (#31)
- Qwen3-Coder 480B-A35B Instruct: 42.0 (#126)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1315 | 1379 |
| Fiction.LiveBench | 77.8% | — |

## Writing & Preference

- o4-mini: 54.2 (#148)
- Qwen3-Coder 480B-A35B Instruct: 55.5 (#142)

| Benchmark | o4-mini | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1353 | 1356 |
| LMArena Creative Writing | 1294 | 1333 |
| LMArena Multi-Turn | 1350 | 1365 |
| Short-Story Creative Writing | 75% | — |
| WildBench | 85.4% | — |

## FAQ

### Is o4-mini better than Qwen3-Coder 480B-A35B Instruct?

o4-mini is the stronger model overall, scoring 41.6 to 37.7 on the Noometry Index.

### Which is cheaper, o4-mini or Qwen3-Coder 480B-A35B Instruct?

o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.

### Is o4-mini or Qwen3-Coder 480B-A35B Instruct better for coding?

o4-mini scores higher on coding benchmarks: 41.6 versus 36.2 in the Noometry coding category.

### Which has the bigger context window?

Qwen3-Coder 480B-A35B Instruct does, with 262K tokens against 200K.

### How many benchmarks do o4-mini and Qwen3-Coder 480B-A35B Instruct share?

23 benchmarks have published results for both models. o4-mini has 60 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.