MicroLLM Lab – Try 7 tiny LLM's in the browser

MicroLLM Lab is an in-browser tool that allows users to test and benchmark seven small language models using WebGPU, measuring their performance metrics like tokens per second and accuracy on objective tests. The platform enables users to run performance comparisons, generate verifiable benchmark certificates with their hardware specifications, and create custom JavaScript benchmarks to evaluate model outputs. All data and results are stored locally on the user's device via browser cache.