Unverified50% confidenceTradeoffExact time
Audio is more sensitive to numerical precision and distortion is easily detected by the ear, so fp32 is chosen for the audio VAE; fp16 is chosen for the video VAE to strike a balance between image quality and VRAM usage.
1
Sources
50%
Confidence
Long-term
Relevance
9/16/2026
First Seen
Sources
Related Entities
Cite This Claim
Stable URI
https://kongchang.com/claim/924392API
curl https://kongchang.com/api/v1/knowledge/claims/924392MCP
get_claim(id=924392)