Falcon-H1-3B-Instruct: benchmark results
Provider: TII. Released 2025-05-20. Access: Open.
Aggregate Rating 1365 ± 34, rank #1317 of 1635 rated models, from 67 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Granite Arabic Card - Question Answering - Belebele (En) | 90.44 | Exact Match (%) | 88.9 |
| Granite Arabic Card - Question Answering - DialArMMLU (En) | 71.48 | Exact Match (%) | 88.9 |
| Granite Arabic Card - Question Answering - MMLU-ProX (En) | 52.06 | Exact Match (%) | 88.9 |
| Open Portuguese LLM - FaQuAD NLI | 79.67 | Macro F1 (%) | 87.7 |
| Open Portuguese LLM - BLUEX | 61.89 | Accuracy (%) | 81.3 |
| Granite Arabic Card - Question Answering - Belebele (MSA) | 80 | Exact Match (%) | 77.8 |
| Open Portuguese LLM - ENEM | 69.07 | Accuracy (%) | 72.1 |
| Open Portuguese LLM - ASSIN2 RTE | 91.9 | Macro F1 (%) | 71.8 |
| Open Portuguese LLM - OAB Exams | 48.56 | Accuracy (%) | 70 |
| Granite Arabic Card - Question Answering - Average | 61.58 | Exact Match (%) | 55.6 |
| Granite Arabic Card - Question Answering - Global PIQA (Dial) | 70.8 | Exact Match (%) | 55.6 |
| Granite Arabic Card - Question Answering - Global PIQA (MSA) | 66 | Exact Match (%) | 55.6 |
Interactive version: theaggregate.ai/model?slug=falcon-h1-3b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-05.