Generative Tasks Evaluation of Arabic LLMs
Generate relation triplets from text
Detect harms and risks with Granite Guardian 3.1 8B
Playground for NuExtract-v1.5
This is for learning purpose, don't take it seriously :)
Retrieve news articles based on a query
Humanize AI-generated text to sound like it was written by a human
Analyze text to identify entities and relationships
Parse and highlight entities in an email thread
Search for philosophical answers by author
Classify Turkish news into categories
Explore and filter language model benchmark results
Track, rank and evaluate open LLMs and chatbots
AraGen Leaderboard is a platform designed for evaluating and comparing the performance of Arabic language models (LLMs) on generative tasks. It serves as a benchmarking tool to assess the capabilities of different models in understanding and generating Arabic text. AraGen Leaderboard is free to use and is particularly useful for researchers, developers, and users interested in Arabic NLP.
What is AraGen Leaderboard used for?
AraGen Leaderboard is used to evaluate and compare the performance of Arabic language models on generative tasks, helping users identify the best models for their needs.
How are models evaluated on AraGen Leaderboard?
Models are evaluated based on predefined metrics and custom tasks set by the user, ensuring a comprehensive assessment of their capabilities.
Can I use AraGen Leaderboard for languages other than Arabic?
No, AraGen Leaderboard is specifically designed for evaluating Arabic language models and does not support other languages.