SigLIP2 NaFlex
A family of vision encoders proposed by Google. SigLIP stands for Sigmoid Loss for Language-Image Pre-training, replacing CLIP's softmax contrastive loss with a sigmoid loss. NaFlex (Native Flexible Resolution) supports processing images at their native aspect ratio and variable resolution.
Timeline (last 90 days)
d1-3B has 3.12B parameters, its vision encoder is a shape-optimized version of SigLIP2 NaFlex (400M), its context length is 32,768 tokens, its vocabulary is 128,000, its weight format is Safetensors/BF16, its base model is LFM2.5-VL-3B (the text backbone was weight-averaged with LFM2.5-2.6B), and its license is lfm1.0.
SigLIP2 NaFlex is a vision encoder family proposed by Google that uses a sigmoid loss and supports processing images at their native aspect ratio and variable resolution.
All Facts (2)
d1-3B has 3.12B parameters, its vision encoder is a shape-optimized version of SigLIP2 NaFlex (400M), its context length is 32,768 tokens, its vocabulary is 128,000, its weight format is Safetensors/BF16, its base model is LFM2.5-VL-3B (the text backbone was weight-averaged with LFM2.5-2.6B), and its license is lfm1.0.
50%UnverifiedSigLIP2 NaFlex is a vision encoder family proposed by Google that uses a sigmoid loss and supports processing images at their native aspect ratio and variable resolution.
50%