Tiny-ML Leaderboard is an interactive web app for browsing sub-150M-parameter language models and comparing their results on benchmarks such as BLiMP, ARC-Easy, and WikiText-2. It also presents efficiency scores for researchers and developers evaluating compact models.
Tiny-ML Leaderboard sits in PulseGate's Model leaderboards category. It focuses on comparing the benchmark performance and efficiency of small language models. Tiny-ML Leaderboard is a B2B product aimed at machine learning researchers and developers. Tiny-ML Leaderboard is free to use. Tiny-ML Leaderboard is available on the web.
It is developed by Glint-Research. Key capabilities include model leaderboard, benchmark scores, and efficiency scores.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do