Memory-efficient PyTorch optimizers with Adaptive Log-Space quantization, fused CUDA kernels, and Adafactor, CAME, APOLLO support.
machine-learning deep-learning cuda pytorch quantization memory-efficient fine-tuning low-rank optimizers low-bit llm-training 8bit-optimizer adafactor memory-efficient-training optimizer-research apollo-optimizer
-
Updated
Aug 24, 2026 - Python