# Granite 4.0 H Small vs Qwen3-Coder 480B-A35B Instruct

> Qwen3-Coder 480B-A35B Instruct is the stronger model overall, scoring 38.1 to 36.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/ibm-granite-h-small-vs-qwen3-coder-480b-a35b-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 13

## Summary

- They share 13 benchmarks with published results for both. Granite 4.0 H Small scores higher in 1 category and Qwen3-Coder 480B-A35B Instruct in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3-Coder 480B-A35B Instruct leads 55.3 to 43.8.

## Snapshot

| | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 36.5 | 38.1 |
| Rank | 214 | 190 |
| Context | — | 262K |
| Input $/M | — | $1.50 |
| Output $/M | — | $7.50 |
| Weights | Open | Open |

## Coding

- Granite 4.0 H Small: 36.4 (#209)
- Qwen3-Coder 480B-A35B Instruct: 35.5 (#223)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Coding | 1249 | 1412 |
| 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

- Granite 4.0 H Small: —
- Qwen3-Coder 480B-A35B Instruct: 23.9 (#123)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |

## Reasoning

- Granite 4.0 H Small: 24.4 (#163)
- Qwen3-Coder 480B-A35B Instruct: 25.5 (#149)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1240 | 1372 |
| Kagi LLM Benchmark | — | 49.5% |

## Math

- Granite 4.0 H Small: 31.4 (#223)
- Qwen3-Coder 480B-A35B Instruct: 37.6 (#150)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1247 | 1365 |
| Omni-MATH | 29.6% | — |

## Knowledge

- Granite 4.0 H Small: 32.0 (#215)
- Qwen3-Coder 480B-A35B Instruct: 37.0 (#162)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1251 | 1338 |
| MMLU-Pro | 56.9% | — |
| Vectara Hallucination Rate | 5.2% | — |
| GPQA (HELM) | 38.3% | — |

## Multilingual

- Granite 4.0 H Small: 38.6 (#228)
- Qwen3-Coder 480B-A35B Instruct: 47.7 (#148)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1216 | 1346 |
| LMArena Chinese | 1249 | 1357 |
| LMArena Russian | 1201 | 1366 |
| LMArena Spanish | 1258 | 1360 |
| LMArena French | — | 1398 |
| LMArena German | — | 1325 |
| LMArena Japanese | — | 1310 |
| LMArena Korean | — | 1305 |

## Instruction Following

- Granite 4.0 H Small: 68.7 (#183)
- Qwen3-Coder 480B-A35B Instruct: 71.6 (#147)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1222 | 1355 |
| IFEval | 89% | — |

## Long Context

- Granite 4.0 H Small: 37.7 (#213)
- Qwen3-Coder 480B-A35B Instruct: 42.0 (#131)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1242 | 1378 |

## Writing & Preference

- Granite 4.0 H Small: 43.8 (#227)
- Qwen3-Coder 480B-A35B Instruct: 55.3 (#147)

| Benchmark | Granite 4.0 H Small | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1241 | 1357 |
| LMArena Creative Writing | 1211 | 1333 |
| LMArena Multi-Turn | 1242 | 1365 |
| WildBench | 73.9% | — |

## FAQ

### Is Granite 4.0 H Small better than Qwen3-Coder 480B-A35B Instruct?

Qwen3-Coder 480B-A35B Instruct is the stronger model overall, scoring 38.1 to 36.5 on the Noometry Index.

### Is Granite 4.0 H Small or Qwen3-Coder 480B-A35B Instruct better for coding?

They score almost the same on coding (36.4 vs 35.5); test both on your own repository before choosing.

### How many benchmarks do Granite 4.0 H Small and Qwen3-Coder 480B-A35B Instruct share?

13 benchmarks have published results for both models. Granite 4.0 H Small has 19 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.