# Granite-4.0-H-1B Benchmark Scores & Performance

> Granite-4.0-H-1B by IBM scores 30.47/100 overall, ranking #165 out of 512 AI models.

Canonical page: https://benchlm.ai/models/granite-4-0-h-1b

Last updated: September 28, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | IBM |
| Source Type | Open Weight |
| Reasoning Type | Non-Reasoning |
| Context Window | 128K |
| Overall Score | 30.47/100 |
| Overall Rank | #165 of 512 |

## Family & Coverage

- Family: Granite 4.0 1B
- Variant: hybrid
- Benchmarks covered: 10 of 486
- Sibling models: [Granite-4.0-1B](/models/granite-4-0-1b)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [τ²-bench results](/benchmarks/tau2-bench) | 19.6% |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-LCR](/benchmarks/lcr) | 7.0% |
| [CritPt](/benchmarks/critpt) | 0.0% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 5.2% |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 26.3% |
| [AA-HLE](/benchmarks/aahle) | 5.0% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | -72.3% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 5.2% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 81.7% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-IFBench](/benchmarks/aaifbench) | 26.2% |

## Other IBM Models

- [Granite 4.2 8B](/models/granite-4-2-8b) - Score: 30.27
- [Granite-4.0-350M](/models/granite-4-0-350m) - Score: 30.21
- [Granite-4.0-H-350M](/models/granite-4-0-h-350m) - Score: 30.21
- [Granite 4.2 30B](/models/granite-4-2-30b) - Score: not computed
- [Granite 4.2 3B](/models/granite-4-2-3b) - Score: not computed
- [Granite Speech 5.0 470M TurboCTC](/models/granite-speech-5-0-470m-turboctc) - Score: not computed
- [Granite-4.0-1B](/models/granite-4-0-1b) - Score: not computed