Leaderboard of information retrieval models in French
Display and submit LLM benchmarks
Browse and submit LLM evaluations
View and submit LLM benchmark evaluations
Display and submit language model evaluations
Display and filter leaderboard models
Rank machines based on LLaMA 7B v2 benchmark results
Measure execution times of BERT models using WebGPU and WASM
View RL Benchmark Reports
Display model benchmark results
Evaluate model predictions with TruLens
Merge Lora adapters with a base model
Compare model weights and visualize differences
DécouvrIR is a leaderboard platform designed to benchmark and compare information retrieval models specifically for the French language. It provides a comprehensive overview of model performance, enabling researchers, developers, and users to understand the effectiveness of various models in retrieving relevant information. The platform is tailored for the French language, making it a valuable resource for those working with French datasets or applications.
• Leaderboard System: Displays benchmark results for multiple information retrieval models in French.
• French Language Support: Specialized for models and datasets in the French language.
• Model Comparison: Allows users to compare performance metrics of different models.
• Regular Updates: The leaderboard is updated with the latest models and results.
• Interactive Filters: Users can filter and sort models based on specific criteria.
• Detailed Metrics: Provides in-depth performance metrics for each model.
What is DécouvrIR?
DécouvrIR is a benchmarking platform for information retrieval models in the French language. It provides a leaderboard to compare model performance and view detailed metrics.
How often is the leaderboard updated?
The leaderboard is regularly updated with new models and results to reflect the latest advancements in information retrieval for French.
Can I use DécouvrIR for non-French models?
No, DécouvrIR is specifically designed for French language models and datasets. It focuses on benchmarking performance in French information retrieval tasks.