# Granite 4.0 H Small vs Trinity Large Thinking

> Trinity Large Thinking is the stronger model overall, scoring 38.1 to 35.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/ibm-granite-h-small-vs-trinity-large-thinking
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. Granite 4.0 H Small scores higher in 2 categories and Trinity Large Thinking in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Trinity Large Thinking leads 54.0 to 44.0.

## Snapshot

| | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| Provider | IBM | Arcee AI |
| Noometry Index | 35.9 | 38.1 |
| Rank | 213 | 180 |
| Context | — | 262K |
| Input $/M | — | $0.25 |
| Output $/M | — | $0.80 |
| Weights | Open | Open |

## Coding

- Granite 4.0 H Small: 35.8 (#207)
- Trinity Large Thinking: 33.8 (#238)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Coding | 1247 | 1381 |
| LMArena WebDev | — | 1238 |
| SciCode | — | 36.1% |

## Reasoning

- Granite 4.0 H Small: 23.4 (#164)
- Trinity Large Thinking: 16.3 (#298)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Hard Prompts | 1240 | 1350 |
| NYT Connections (extended) | — | 16.5% |
| CritPt | — | 0.9% |
| Thematic Generalization | — | 41.6% |
| Surface Evolver Bench | — | 15.6% |

## Math

- Granite 4.0 H Small: 29.3 (#219)
- Trinity Large Thinking: 34.9 (#149)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1251 | 1366 |
| Omni-MATH | 29.6% | — |

## Knowledge

- Granite 4.0 H Small: 31.5 (#210)
- Trinity Large Thinking: 40.3 (#107)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 5.2% | 6.9% |
| LMArena Expert | 1244 | 1359 |
| MMLU-Pro | 56.9% | — |
| GPQA (HELM) | 38.3% | — |

## Multilingual

- Granite 4.0 H Small: 38.7 (#223)
- Trinity Large Thinking: 46.4 (#155)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1216 | 1325 |
| LMArena Chinese | 1250 | 1372 |
| LMArena Russian | 1198 | 1338 |
| LMArena Spanish | 1258 | 1356 |
| LMArena French | — | 1374 |
| LMArena German | — | 1356 |
| LMArena Japanese | — | 1312 |
| LMArena Korean | — | 1304 |

## Instruction Following

- Granite 4.0 H Small: 68.9 (#178)
- Trinity Large Thinking: 70.7 (#157)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1221 | 1334 |
| IFEval | 89% | — |

## Long Context

- Granite 4.0 H Small: 37.6 (#209)
- Trinity Large Thinking: 41.2 (#139)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1240 | 1355 |

## Writing & Preference

- Granite 4.0 H Small: 44.0 (#222)
- Trinity Large Thinking: 54.0 (#152)

| Benchmark | Granite 4.0 H Small | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1241 | 1340 |
| LMArena Creative Writing | 1210 | 1321 |
| LMArena Multi-Turn | 1243 | 1342 |
| WildBench | 73.9% | — |

## FAQ

### Is Granite 4.0 H Small better than Trinity Large Thinking?

Trinity Large Thinking is the stronger model overall, scoring 38.1 to 35.9 on the Noometry Index.

### Is Granite 4.0 H Small or Trinity Large Thinking better for coding?

Granite 4.0 H Small scores higher on coding benchmarks: 35.8 versus 33.8 in the Noometry coding category.

### How many benchmarks do Granite 4.0 H Small and Trinity Large Thinking share?

14 benchmarks have published results for both models. Granite 4.0 H Small has 19 scored results on Noometry and Trinity Large Thinking has 24.