Model comparison
Granite 4.0 H Small vs Step 3
Step 3 is the stronger model overall, scoring 39.9 to 35.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
StepFunSummary
- They share 13 benchmarks with published results for both. Granite 4.0 H Small scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3 leads 54.5 to 44.0.
Side by side
| Granite 4.0 H Small | Step 3 | |
|---|---|---|
| Provider | IBM | StepFun |
| Noometry Index | 35.9 | 39.9 |
| Released | — | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 19 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Granite 4.0 H Small: 35.8 (#207), Step 3: 39.5 (#144)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Coding | 1247 | 1367 |
Reasoning Step 3 leads
Granite 4.0 H Small: 23.4 (#164), Step 3: 27.6 (#102)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1240 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
Math Step 3 leads
Granite 4.0 H Small: 29.3 (#219), Step 3: 34.9 (#148)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Math | 1251 | 1366 |
| Omni-MATH | 29.6% | — |
Knowledge Step 3 leads
Granite 4.0 H Small: 31.5 (#210), Step 3: 36.2 (#161)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Expert | 1244 | 1333 |
| MMLU-Pro | 56.9% | — |
| Vectara Hallucination Rate | 5.2% | — |
| GPQA (HELM) | 38.3% | — |
Multimodal Not comparable
Granite 4.0 H Small: —, Step 3: 34.9 (#81)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |
Multilingual Step 3 leads
Granite 4.0 H Small: 38.7 (#223), Step 3: 46.5 (#154)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Non-English | 1216 | 1327 |
| LMArena Chinese | 1250 | 1396 |
| LMArena Russian | 1198 | 1331 |
| LMArena Spanish | 1258 | 1372 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1271 |
Instruction Following Step 3 leads
Granite 4.0 H Small: 68.9 (#178), Step 3: 70.6 (#159)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1221 | 1332 |
| IFEval | 89% | — |
Long Context Step 3 leads
Granite 4.0 H Small: 37.6 (#209), Step 3: 40.3 (#152)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Longer Query | 1240 | 1326 |
Writing & Preference Step 3 leads
Granite 4.0 H Small: 44.0 (#222), Step 3: 54.5 (#146)
| Benchmark | Granite 4.0 H Small | Step 3 |
|---|---|---|
| LMArena Text | 1241 | 1351 |
| LMArena Creative Writing | 1210 | 1322 |
| LMArena Multi-Turn | 1243 | 1340 |
| WildBench | 73.9% | — |
Frequently asked questions
Is Granite 4.0 H Small better than Step 3?
Step 3 is the stronger model overall, scoring 39.9 to 35.9 on the Noometry Index.
Is Granite 4.0 H Small or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 39.5 versus 35.8 in the Noometry coding category.
How many benchmarks do Granite 4.0 H Small and Step 3 share?
13 benchmarks have published results for both models. Granite 4.0 H Small has 19 scored results on Noometry and Step 3 has 17.