Model comparison
Hy3 vs Granite 4.0 H Small
Hy3 is the stronger model overall, scoring 44.2 to 36.5 on the Noometry Index.
Last verified . 13 shared benchmarks.
TencentSummary
- They share 13 benchmarks with published results for both. Hy3 scores higher in 8 categories and Granite 4.0 H Small in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Hy3 leads 62.2 to 43.8.
Side by side
| Hy3 | Granite 4.0 H Small | |
|---|---|---|
| Provider | Tencent | IBM |
| Noometry Index | 44.2 | 36.5 |
| Released | 2026-07-06 | — |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.13 | — |
| Output $ / M tokens | $0.53 | — |
| Results tracked | 19 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.8 (#63), Granite 4.0 H Small: 36.4 (#209)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Coding | 1464 | 1249 |
| LMArena WebDev | 1508 | — |
Reasoning Hy3 leads
Hy3: 26.1 (#136), Granite 4.0 H Small: 24.4 (#163)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1240 |
| NYT Connections (extended) | 41.2% | — |
Math Hy3 leads
Hy3: 40.1 (#93), Granite 4.0 H Small: 31.4 (#223)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Math | 1475 | 1247 |
| Omni-MATH | — | 29.6% |
Knowledge Hy3 leads
Hy3: 40.8 (#114), Granite 4.0 H Small: 32.0 (#215)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Expert | 1460 | 1251 |
| MMLU-Pro | — | 56.9% |
| Vectara Hallucination Rate | — | 5.2% |
| GPQA (HELM) | — | 38.3% |
Multilingual Hy3 leads
Hy3: 53.5 (#65), Granite 4.0 H Small: 38.6 (#228)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Non-English | 1426 | 1216 |
| LMArena Chinese | 1493 | 1249 |
| LMArena Russian | 1432 | 1201 |
| LMArena Spanish | 1456 | 1258 |
| LMArena French | 1461 | — |
| LMArena German | 1439 | — |
| LMArena Japanese | 1392 | — |
| LMArena Korean | 1395 | — |
Instruction Following Hy3 leads
Hy3: 75.1 (#70), Granite 4.0 H Small: 68.7 (#183)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Instruction Following | 1426 | 1222 |
| IFEval | — | 89% |
Long Context Hy3 leads
Hy3: 44.1 (#75), Granite 4.0 H Small: 37.7 (#213)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Longer Query | 1442 | 1242 |
Writing & Preference Hy3 leads
Hy3: 62.2 (#81), Granite 4.0 H Small: 43.8 (#227)
| Benchmark | Hy3 | Granite 4.0 H Small |
|---|---|---|
| LMArena Text | 1439 | 1241 |
| LMArena Creative Writing | 1402 | 1211 |
| LMArena Multi-Turn | 1436 | 1242 |
| WildBench | — | 73.9% |
Frequently asked questions
Is Hy3 better than Granite 4.0 H Small?
Hy3 is the stronger model overall, scoring 44.2 to 36.5 on the Noometry Index.
Is Hy3 or Granite 4.0 H Small better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 36.4 in the Noometry coding category.
How many benchmarks do Hy3 and Granite 4.0 H Small share?
13 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Granite 4.0 H Small has 19.