| --- |
| library_name: pytorch |
| --- |
| |
|  |
|
|
| **SigLIP 2** is a multilingual vision–language encoder that extends the original SigLIP training objective with captioning, self-distillation, masked prediction, and improved data curation for stronger semantic understanding, localization, and dense visual representations. |
|
|
| Original paper: [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features, Michael Tschannen, et al., 2025](https://arxiv.org/abs/2502.14786) |
|
|
| # SigLIP2-Base-Patch16-224 |
|
|
| This model uses the **SigLIP 2 Base-Patch16-224x224** variant, based on a ViT-Base vision encoder with **16×16 image patches** and approximately **86M parameters**. It is well suited for zero-shot image classification, image–text retrieval, and as a vision encoder for VLMs and downstream vision tasks. |
|
|
| Model Configuration: |
| - Reference implementation: [google/siglip2-base-patch16-224](https://huggingface.co/google/siglip2-base-patch16-224) |
| - Original Weight: [SigLIP2-Base-Patch16](https://huggingface.co/google/siglip2-base-patch16-224) |
| - Dataset: [ImageNet](https://image-net.org) |
| - Resolution: 3x224x224 |
| - Support Cooper version: |
| - Cooper SDK: [2.5.4] |
| - Cooper Foundry: [2.3] |
|
|
| | Model | Device | Model Link | |
| | :-----: | :-----: | :-----: | |
| | SigLIP2-Base-Patch16 Image Encoder | N1-655 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/n1-655_siglip2_base_patch16_image_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Text Encoder | N1-655 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/n1-655_siglip2_base_patch16_text_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Image Encoder | X7 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/x7_siglip2_base_patch16_image_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Text Encoder | X7 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/x7_siglip2_base_patch16_text_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Image Encoder | CV7 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/cv7_siglip2_base_patch16_image_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Text Encoder | CV7 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/cv7_siglip2_base_patch16_text_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Image Encoder | CV72 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/cv72_siglip2_base_patch16_image_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Text Encoder | CV72 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/cv72_siglip2_base_patch16_text_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Image Encoder | CV75 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/cv75_siglip2_base_patch16_image_encoder_act16.bin) | |
| | SigLIP2-Base-Patch16 Text Encoder | CV75 | [Model_Link](https://huggingface.co/Ambarella/SigLIP2/blob/main/cv75_siglip2_base_patch16_text_encoder_act16.bin) | |
|
|
