Spaces:
Running
Running
|
Download README.md from OpenEvals/leaderboard-watcher: direct link, hf CLI and curl.
- Browser
- Download file 1.02 kB
-
https://huggingface.co/spaces/OpenEvals/leaderboard-watcher/resolve/main/README.md
- Command line
-
hf download hf://spaces/OpenEvals/leaderboard-watcher/README.md
-
curl -L -o README.md https://huggingface.co/spaces/OpenEvals/leaderboard-watcher/resolve/main/README.md
1.02 kB
A newer version of the Gradio SDK is available: 6.31.0
metadata
title: HF Hub Benchmark Dashboard
emoji: 🏆
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 5.29.0
app_file: app.py
pinned: false
short_description: Live dashboard for HF Hub benchmark leaderboards
HF Hub Benchmark Dashboard
Live dashboard for every official benchmark leaderboard published on the Hugging Face Hub.
Features
- Sidebar navigation by category (Knowledge, Math, Code, Agents, Vision, …)
- Per-benchmark leaderboard with rank medals, model links, and provider chips
- Inference provider enrichment from router.huggingface.co: price/1M tokens (in & out), TTFT, throughput, context length
- License and parameter count per model (from the HF models API)
- Filter to only show models with live inference providers
- 6-hour cache with manual refresh
HF Token
Set HF_TOKEN as a Space secret to access gated benchmarks (e.g. HLE).
Public benchmarks work without a token.
Local run
pip install gradio>=5.0.0
python app.py