Radar Topics Builders Weekly Reads

Open Source Radar

Topic: 4bit

← Back to the radar

CategoriesAgents MCP Coding agents LLM RAG Inference Vector DBs Browser Workflows topic: 4bit ✕

SortTrending Most stars Newest1 repos · page 1/1

IST-DASLab/

FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.

1.1k90Pythonquiet