Open-source AI project: SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime.
Direct answer
neural-compressor is best for ai models. Source-backed open-source listing awaiting editorial review.
Trust signals
Dera’s take
Use neural-compressor to turn into .
Source-backed open-source listing awaiting editorial review. Best for AI Models.
Skip it if you need more control, a lower price, or a different output style.
Tasks where this tool may help.
We are still adding guides for this tool. Try a related task or check back later.
FAQ