# Granite 4.0 H Small vs Qwen Max

> Granite 4.0 H Small is the stronger model overall, scoring 36.5 to 34.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/ibm-granite-h-small-vs-qwen-max
- 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 4 categories and Qwen Max in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Granite 4.0 H Small leads 31.4 to 22.3.
- Granite 4.0 H Small has downloadable open weights; the other is API-only.

## Snapshot

| | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 36.5 | 34.7 |
| Rank | 214 | 230 |
| Context | — | 33K |
| Input $/M | — | $1.60 |
| Output $/M | — | $6.40 |
| Weights | Open | Proprietary |

## Coding

- Granite 4.0 H Small: 36.4 (#209)
- Qwen Max: 30.7 (#292)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Coding | 1249 | 1288 |
| Aider Polyglot | — | 21.8% |

## Reasoning

- Granite 4.0 H Small: 24.4 (#163)
- Qwen Max: 25.1 (#151)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | 1240 | 1269 |

## Math

- Granite 4.0 H Small: 31.4 (#223)
- Qwen Max: 22.3 (#276)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Math | 1247 | 1275 |
| OTIS Mock AIME 2024-2025 | — | 16.1% |
| Omni-MATH | 29.6% | — |
| MATH Level 5 | — | 67.2% |
| FrontierMath (Feb 2025 set) | — | 1% |

## Knowledge

- Granite 4.0 H Small: 32.0 (#215)
- Qwen Max: 30.3 (#228)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Expert | 1251 | 1248 |
| GPQA Diamond | — | 56.1% |
| MMLU-Pro | 56.9% | — |
| Vectara Hallucination Rate | 5.2% | — |
| GPQA (HELM) | 38.3% | — |

## Multilingual

- Granite 4.0 H Small: 38.6 (#228)
- Qwen Max: 41.8 (#202)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Non-English | 1216 | 1263 |
| LMArena Chinese | 1249 | 1254 |
| LMArena Russian | 1201 | 1274 |
| LMArena Spanish | 1258 | 1290 |
| LMArena French | — | 1330 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1205 |
| LMArena Korean | — | 1142 |

## Instruction Following

- Granite 4.0 H Small: 68.7 (#183)
- Qwen Max: 66.5 (#208)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Instruction Following | 1222 | 1262 |
| IFEval | 89% | — |

## Long Context

- Granite 4.0 H Small: 37.7 (#213)
- Qwen Max: 39.4 (#180)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Longer Query | 1242 | 1288 |
| Fiction.LiveBench | — | 66.7% |

## Writing & Preference

- Granite 4.0 H Small: 43.8 (#227)
- Qwen Max: 47.8 (#205)

| Benchmark | Granite 4.0 H Small | Qwen Max |
|---|---|---|
| LMArena Text | 1241 | 1282 |
| LMArena Creative Writing | 1211 | 1248 |
| LMArena Multi-Turn | 1242 | 1277 |
| WildBench | 73.9% | — |

## FAQ

### Is Granite 4.0 H Small better than Qwen Max?

Granite 4.0 H Small is the stronger model overall, scoring 36.5 to 34.7 on the Noometry Index.

### Is Granite 4.0 H Small or Qwen Max better for coding?

Granite 4.0 H Small scores higher on coding benchmarks: 36.4 versus 30.7 in the Noometry coding category.

### How many benchmarks do Granite 4.0 H Small and Qwen Max share?

13 benchmarks have published results for both models. Granite 4.0 H Small has 19 scored results on Noometry and Qwen Max has 23.