Verified65% confidenceFactTime unknown
Stable Diffusion's text guidance converts prompts into vectors through the CLIP text encoder.
3
Sources
65%
Confidence
Long-term
Relevance
6/2/2026
First Seen
Sources
Related Entities
Related Claims
Unverified在Stable Diffusion中，CLIP文本编码器将自然语言切分为最多77个token，每个token映射为768维或1024维嵌入向量79% similarUnverified早期Stable Diffusion系列以CLIP为文本编码器，对超过77个Token的提示词处理能力受限79% similarUnverifiedThe core principle of Stable Diffusion is to perform denoising generation in a compressed latent space, using a three-stage process consisting of a CLIP text encoder to understand prompts, a U-Net architecture to perform denoising, and a VAE decoder to restore the image.78% similarVerifiedText instructions are converted into semantic vectors through cross-modal encoders such as CLIP, embedding text and images into the same semantic space to guide image generation.76% similarUnverifiedCLIP is the most commonly used text encoder for the Stable Diffusion series of models, was proposed by OpenAI, and maps images and text to the same semantic space through contrastive learning76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/36410API
curl https://kongchang.com/api/v1/knowledge/claims/36410MCP
get_claim(id=36410)