Skip to main content
← Back to sources

QIMMA: Quality-First Arabic LLM Evaluation Leaderboard Launched by TII

Published 2026-04-21Foundation ModelsLow⭐ Timeline Candidate

Summary

The Technology Innovation Institute (TII), based in the UAE, has launched QIMMA (قِمّة), a new leaderboard on Hugging Face designed to evaluate large language models specifically on Arabic language capabilities. The initiative aims to provide a quality-first benchmarking framework for Arabic NLP, addressing the gap in rigorous evaluation standards for non-English LLMs. The leaderboard is hosted on Hugging Face and authored by researchers including Leen AlQadi and Ahmed Alzubaidi. QIMMA represen

Alignment: Neutral
Related Positions: multi-model-multi-vendor.md
arabic-nlpllm-evaluationleaderboardhugging-facetiimultilingual-aifoundation-modelsmodel-benchmarkingnon-english-llm
QIMMA: Quality-First Arabic LLM Evaluation Leaderboard Launched by TII — Intelligence — Agentic Developer Tools Radar · Signal