Model reference · open weights
siglip-patch16-224 is an open-weight embedding model from google, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
SigLIP (base-sized model) SigLIP model pre-trained on WebLi at resolution 224x224. It was introduced in the paper Sigmoid Loss for Language Image Pre-Training by Zhai et al. and first released in this repository. Disclaimer: The team releasing SigLIP did not write a model card for this model so this model card has been written by the Hugging Face team. Model description SigLIP is CLIP, a multimodal model, with a better loss function. The sigmoid loss operates solely on image-text pairs and does not require a global view of the pairwise similarities for normalization. This allows further scaling up the batch size, while also performing better at smaller batch sizes. A TLDR of SigLIP by one of the authors can be found here. Intended uses & limitations You can use the raw model for tasks like zero-shot image classification and image-text retrieval. See the model hub to look for other versions on a task that interests you. How to use Here is how to use this model to perform zero-shot image classification: Alternatively, one can leverage the pipeline API which abstracts away the complexity for the user: For more code examples, we refer to the documentation. Training procedure Training data SigLIP is pre-trained on the English image-text pairs of the WebLI dataset (Chen et al., 2023). Preprocessing Images are resized/rescaled to the same resolution (224x224) and normalized across the RGB channels with mean (0.5, 0.5, 0.5) and standard deviation (0.5, 0.5, 0.5). Texts are tokenized and padded to the same length (64 tokens). Compute The model was trained on 16 TPU-v4 chips for three days. Evaluation results Evaluation of SigLIP compared to CLIP is shown below (taken from the paper). alt="drawing" width="600"/ BibTeX entry and citation info
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | |
|---|---|
| Type | Embedding models |
| Parameters (lead) | 203M |
| Variants | 1 |
| Runs with | transformers |
| Released | 2023-09-30 |
| Popularity | 1.5M downloads / month |
| Likes | 89 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| siglip-base-patch16-224 | 203M | BF16 | ~0.5 GB | ✓ | Weights ↗ |
Using it via the API
Once AxForge deploys siglip-patch16-224 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (siglip-patch16-224 below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"siglip-patch16-224","input":"text to embed"}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore