Verified65% confidenceFactTime unknown
The Gemma series of models uses the SigLIP vision encoder to process image inputs.
3
Sources
65%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
localOCR：本地部署的开源OCR方案，支持Gemma-4等视觉模型
githubCuriosity-Ai-BV
Related Entities
Related Claims
UnverifiedGemma 4支持完全离线状态下的图像文字提取（OCR），采用视觉-语言多模态架构75% similarUnverifiedThe Gemma 3 series supports understanding and generation of text, images, audio, and video, and its smaller model sizes (such as the 4B and 12B parameter versions) can run on consumer-grade GPUs.70% similarUnverifiedGemma 4 on Apple Silicon now automatically selects the optimal resolution for each image to preserve more detail.67% similarUnverifiedThe new model has gained native visual input capability, enabling it to read visual inputs and access a browser to 'see' its own outputs and iterate.67% similarUnverifiedThe visual model receives each screenshot and outputs structured information, including OCR-recognized text, page layout, pixel coordinate bounding boxes of UI controls, and element states.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/29139API
curl https://kongchang.com/api/v1/knowledge/claims/29139MCP
get_claim(id=29139)