Fanar-C-2-27B: benchmark results

Provider: Other. Access: Open.

Unified ELO 1648 ± 41, rank #463 of 2656 rated models, from 24 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Arabic Broad Leaderboard - RAG QA7.81Average Score (0-10)77.7
BALSAM - Creative Writing50.68Overall score (0-100, LLM-judged generation and multiple cho71.4
Arabic Broad Leaderboard - General Knowledge8.86Average Score (0-10)69.1
BALSAM - Question Answering70.35Overall score (0-100, LLM-judged generation and multiple cho67.9
BALSAM - Information Extraction49.34Overall score (0-100, LLM-judged generation and multiple cho64.3
BALSAM - Translation/Transliteration63.16Overall score (0-100, LLM-judged generation and multiple cho64.3
BALSAM - Factuality22.11Overall score (0-100, LLM-judged generation and multiple cho60.7
BALSAM - Sequence Tagging37.11Overall score (0-100, LLM-judged generation and multiple cho59.3
BALSAM - Fill in the Blank75.46Overall score (0-100, LLM-judged generation and multiple cho53.6
BALSAM - Text Manipulation60.57Overall score (0-100, LLM-judged generation and multiple cho53.6
BALSAM - Overall53.17Mean of category overall scores (0-100)51.9
Arabic Broad Leaderboard - Reasoning & Math7.61Average Score (0-10)51.4

Interactive version: theaggregate.ai/model?slug=fanar-c-2-27b · How It Works · Data refreshed daily, snapshot 2026-09-19.