Models Hardware Evals Leaderboard

Models Leaderboard Hardware Marketplace Evals Train Rentals API Docs

Language

Lucebox Hub campaign banner

LocalMaxxingCommunity benchmarks for local LLM inference.·MIT licensed

Leaderboard Models Hardware Evals Marketplace Rentals Train API Docs Submit GitHub

Back to leaderboard

Submit a benchmark

Share your local LLM inference results with reproducible hardware, engine, and runtime details.

Model

—

Hardware

—

Tok/s

—

Step 1

Choose the model

Enter a Hugging Face model ID, verify it exists, and optionally pin the exact revision for reproducibility.

HuggingFace model ID*

Model revision / commit hash

Trending now