70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float Paper • 2504.11651 • Published 11 days ago • 24
Ultra Collection A series of extremely efficient and high quality custom-made models, including a prototype for a simulated AGI. • 2 items • Updated 9 days ago • 1
Phi-3 Collection Phi-3 family of small language and multi-modal models. Language models are available in short- and long-context lengths. • 26 items • Updated 9 days ago • 566