IBMGranite 4.0 H Small
Granite 4.0 H Small is an IBM open-weights model, released 22 Sept 2025. Its best configuration ranks #642 of 785 on the BenchLeader Index at 41.0 ±3.2, in the lower half. The ± is the point: 275 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in maths (44) and lowest in long context (29). Last measured 22 Oct 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- 128k
- Full answer
- –
- Cost per run
- –
- Released
- 22 Sept 2025
- Overall index41
- Reasoning42
- Coding44
- Agents & tools36
- Maths44
- Knowledge34
- Instruction following35
- Human preference44
- Long context29
- Composite36
How its index has moved
9 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1304#259 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 41.6%#471 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 3.8%#564 | Artificial Analysis | |
| CritPt | 0.0%#315 | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1330#258 | Arena (LMArena) |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 2.3%#320 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 17.3%#351 | Artificial Analysis | |
| BrowseCompnot in index | 0.4%#32 | OpenAI |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Maths | 1281#244 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-Omniscience | -61.0#470 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 14.3%#440 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 12.2%#379 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 31.5%#351 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1270#266 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1288#266 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1242#271 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1279#258 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1293#260 | Arena (LMArena) |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 11.3%#463 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 6.0#543 | Artificial Analysis |
More from IBM
See also
Questions people ask
- What are Granite 4.0 H Small's benchmark scores?
- As of 24 Oct 2026, Granite 4.0 H Small scores 41.0 ±3.2 on the BenchLeader Index, strongest in maths (44) and weakest in long context (29). That rests on 10 independent benchmark results, and every score on this page links to its publisher.
- How does Granite 4.0 H Small rank?
- Granite 4.0 H Small is #642 of 785 ranked configurations on 24 Oct 2026. 275 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- When was Granite 4.0 H Small released?
- IBM released Granite 4.0 H Small on 22 Sept 2025.
Data as of 24 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Granite 4.0 H Small: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/granite-4-0-h-small.