vLLM Quantization Documentation (docs.vllm.ai)

vLLM quantization documentation. Contains the Implementation x Hardware support table showing which quantization methods work on which GPU architectures. Documents --linear-backend and --kernel-config options for selecting quantization backends. Shows Marlin supports Turing (SM7.5) and above but not Volta (SM7.0). Lists supported formats: AutoAWQ, BitsAndBytes, GPTQModel, Intel Neural Compressor, LLM Compressor (FP8, INT8, INT4), NVIDIA Model Optimizer, Online Quantization, AMD Quark, Quantized KV Cache, TorchAO.