Unverified50% confidenceFactExact time
The 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.
1
Sources
50%
Confidence
Long-term
Relevance
9/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe Gemma 3 series released in 2025 uses a SigLIP vision encoder jointly trained with a language model, supports image understanding, and extends the context window to 128K tokens.72% similarUnverifiedThe Gemma series of models uses the SigLIP vision encoder to process image inputs.70% similarUnverifiedGemma 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69% similarUnverifiedGemma 4 E2B 支持三大离线多模态能力：AI Chat（离线对话）、Ask Image（拍照识别）、Ask Audio（本地音频转录）66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/838874API
curl https://kongchang.com/api/v1/knowledge/claims/838874MCP
get_claim(id=838874)