MoonlitGitHubRepoMoonlitThis is a collection of our research on efficient AI, covering hardware-aware NAS and model compression.inference-efficiencymodel-compressionneural-architecture-search91PythonMIT2 y agomicrosoftShare or keepTake it
ParityLanguage and translationDemoParityCut non-English token cost, with a certified boundstatictokenizervocabulary-adaptationRuns in the browser on Hugging Face0apache-2.07 d agoNNagaYuShare or keepTry it