Fanar-C-2-27B: benchmark results
Provider: QCRI. Access: API.
Aggregate Rating 1541 ± 21, rank #637 of 1635 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BALSAM - Creative Writing | 50.68 | Overall score (0-100, LLM-judged generation and multiple cho | 78.8 |
| Arabic Broad Leaderboard - RAG QA | 7.81 | Average Score (0-10) | 75.4 |
| BALSAM - Information Extraction | 49.34 | Overall score (0-100, LLM-judged generation and multiple cho | 74.3 |
| BALSAM - Factuality | 22.11 | Overall score (0-100, LLM-judged generation and multiple cho | 73 |
| BALSAM - Question Answering | 70.35 | Overall score (0-100, LLM-judged generation and multiple cho | 73 |
| Arabic Broad Leaderboard - General Knowledge | 8.86 | Average Score (0-10) | 67.5 |
| BALSAM - Fill in the Blank | 75.46 | Overall score (0-100, LLM-judged generation and multiple cho | 66.7 |
| BALSAM - Reading Comprehension | 52.34 | Overall score (0-100, LLM-judged generation and multiple cho | 66.7 |
| BALSAM - Sequence Tagging | 37.11 | Overall score (0-100, LLM-judged generation and multiple cho | 65.7 |
| BALSAM - Overall | 53.17 | Mean of category overall scores (0-100) | 58.6 |
| BALSAM - Translation/Transliteration | 63.16 | Overall score (0-100, LLM-judged generation and multiple cho | 54.1 |
| BALSAM - Text Manipulation | 60.57 | Overall score (0-100, LLM-judged generation and multiple cho | 50 |
Interactive version: theaggregate.ai/model?slug=fanar-c-2-27b · How It Works · Data refreshed daily, snapshot 2026-11-05.