IST-DASLab/marlin · AI Repo StatsISIST-DASLab/marlin
FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.
LLM & mô hình nền tảngPython Apache-2.0 98tạo 2024-01-17 #4bit#kernel#llm#quantization