Unverified60% confidenceFactExact time
CLIP contains two parallel encoders: the text encoder uses a Transformer-based architecture (a 12-layer Transformer similar to GPT-2), and the visual encoder has two variants—one based on ResNet and one based on ViT.
2
Sources
60%
Confidence
Long-term
Relevance
8/5/2026
First Seen
Sources
Related Claims
UnverifiedCLIP adopts a dual-tower architecture, consisting of a pretrained visual encoder (such as ViT) and a pretrained text encoder.76% similarUnverifiedDINOv2使用Vision Transformer（ViT）架构，将输入图像划分为固定大小的补丁，每个补丁经Transformer编码后产生高维嵌入向量70% similarVerifiedGPT-Image-2 very likely adopts a native multimodal architecture in which text understanding and image generation share the same Transformer backbone network.67% similarUnverifiedMainstream VLMs typically use a pretrained Vision Transformer (ViT) as the visual encoder, such as CLIP ViT and SigLIP.67% similarUnverifiedQwen VL的视觉编码器基于ViT（Vision Transformer）架构，将图像像素信息压缩编码为语言模型可理解的token序列67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/691015API
curl https://kongchang.com/api/v1/knowledge/claims/691015MCP
get_claim(id=691015)