Update UI: Close filters by default and increase quick filter button size

Migrate to Gradio app with interactive features

Remove authentication requirement - dataset is public

Update website to load from parquet dataset

Update fetch script to upload parquet dataset to HuggingFace

Add UV script metadata to fetch_api_only.py

Fetch leaderboard data from private HuggingFace dataset

Update default page state: deselect GSM8K and sort by TB2

Match HuggingFace design system and remove About section

Make column sorting more obvious and accessible

Implement benchmark column sorting

Clean up repository - remove unused files

Remove progress bars under benchmark scores

Fix sticky header alignment to match table position

Implement JavaScript-based sticky table header

Fix sticky table header to stay visible when scrolling

Use more descriptive benchmark column names

Remove aggregate score column from table

Add SWE-bench Pro and EvasionBench data

Fix: Remove darker background for top models in name column

Remove green background highlight for top models

Implement category-based gradient colors for benchmark scores

Remove verified/confidence badges from score cells

Fix: Restore provider logos

Increase table header size

Fix: Use safetensors.total for accurate parameter counts

Fix: Restore JavaScript and update LEADERBOARD_DATA with parametersInBillions

Update LEADERBOARD_DATA with parametersInBillions field

Debug: Add console logging to size filter

Add model size range slider filter

Move benchmark filter to separate bar below search

Make categories more compact to fit horizontally without scrollbar

Force categories to display in single horizontal row

Make benchmark categories more horizontal and compact

Refresh leaderboard data with latest benchmarks

Improve category differentiation in benchmark filter

Replace dropdown with category-based benchmark filter UI

Update leaderboard data with GPQA and HLE scores

Move provider logo inline with model name

Move coverage count under aggregate score like original

Replace category buttons with benchmark dropdown

Fix aggregate scoring - use original scores instead of recalculating

Add dynamic aggregate score calculation and smart model filtering

Add category filters and remove Open Source button

Fix OAuth scopes in README metadata

Add official benchmarks leaderboard with OAuth support

Update index.html

Update index-BACK2.html

Update index.html

Update index.html