Qalb-1.0-8B-Instruct: benchmark results
Provider: Other.
Aggregate Rating 1372 ± 36, rank #7130 of 10916 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UrduMMLU (Urdu Prompt) - Humanities | 31.39 | Humanities (10,982 questions: ethics, fine arts, Islamic stu | 28.6 |
| UrduMMLU (English Prompt) | 34.82 | Overall exact-match accuracy (%) on UrduMMLU's Urdu multiple | 27.6 |
| UrduMMLU (Urdu Prompt) - General Knowledge | 36.05 | General knowledge (1,362 questions, the Other domain) exact- | 21.4 |
| UrduMMLU (Urdu Prompt) | 31.54 | Overall exact-match accuracy (%) on UrduMMLU's Urdu multiple | 17.9 |
| UrduMMLU (Urdu Prompt) - Profession | 35.46 | Profession (973 questions: professional development and stud | 17.9 |
| UrduMMLU (Urdu Prompt) - STEM | 30.59 | STEM (5,113 questions: biology, chemistry, computer science, | 14.3 |
| UrduMMLU (Urdu Prompt) - Social Sciences | 31.12 | Social Sciences (7,959 questions in 11 subjects, including P | 14.3 |
Interactive version: theaggregate.ai/model?slug=qalb-1-0-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.