vllm-omni/docs/design/module/quantization.md

title, kind, status, owners, primary_code_paths, related_code_paths, depends_on, validation_paths, upstream_refs, last_reviewed

title	kind	status	owners	primary_code_paths	related_code_paths	depends_on	validation_paths	upstream_refs	last_reviewed
Quantization	module	draft	@david6666666 @Isotr0py @lishunyang12	vllm_omni/quantization/** vllm_omni/diffusion/quantization/**	vllm_omni/platforms/*/quant/**	model_integration.md diffusion/diffusion_model_integration.md	tests/quantization/** tests/diffusion/quantization/**	vllm.model_executor.layers.quantization	2026-07-16

Quantization

Quantization owns method registration, checkpoint interpretation, layer replacement, and runtime constraints for reduced-precision execution.

Candidate invariants

QUANT-INV-001: Checkpoint and runtime methods agree

Rule: The selected method MUST validate checkpoint metadata, supported layers, device capabilities, and required dependencies before use.

QUANT-INV-002: Unsupported combinations fail clearly

Rule: Quantization MUST NOT silently fall back to a different precision or method when a requested combination is unsupported.

QUANT-INV-003: Numerical validation is required

Rule: A new method or layer mapping MUST include correctness comparison against a defined reference result.

Safe-change guide

Test loading, representative layers, metadata, platform guards, and numerical tolerances separately from performance.