siglip2-so400m-patch14-384
by google
SigLIP 2 SO400M (patch14, 384px): a top open image-text embedder for zero-shot visual search
google/siglip2-so400m-patch14-384Overview
SigLIP 2 is Google's second-generation image-text encoder. It keeps SigLIP's sigmoid contrastive loss (which scales better than CLIP's softmax across large batches) and adds captioning-based pretraining, self-supervised objectives (self-distillation and masked prediction), and online data curation into one recipe. The result is stronger zero-shot retrieval, noticeably better localization, and denser features than the original SigLIP, with multilingual coverage. This SO400M checkpoint is the shape-optimized ~400M-parameter model at 14px patches and 384px input, a strong quality-per-FLOP point for production.
On Mixpeek, SigLIP 2 is a leading choice for the image and frame embedding stage: it maps pictures and text into one shared space so an agent can search a visual library by description or by example image. Its improved spatial/dense features also help downstream localization and region-level retrieval, not just whole-image matching.
Architecture
Vision-language dual encoder trained with a sigmoid (pairwise) contrastive loss instead of softmax, so it scales to very large batches without a global normalization term. SigLIP 2 augments that with a captioning decoder head, self-supervised self-distillation and masked-prediction losses, and online data curation. SO400M is a shape-optimized ~400M-parameter ViT; the patch14-384 variant uses 14px patches at 384px resolution for finer spatial detail.
Key Capabilities
- •Zero-shot image-text retrieval in a shared embedding space
- •Improved localization and dense features vs SigLIP v1
- •Multilingual text-image alignment
- •Sigmoid loss scales to large-batch training
Use Cases on Mixpeek
- •Embed an image library for text-to-image and image-to-image search
- •Frame embeddings for video visual search and dedup
- •Region/crop retrieval where dense features matter
- •The visual half of a hybrid dense + BM25 retriever
Tags
Use siglip2-so400m-patch14-384 on Mixpeek
Build multimodal processing pipelines with this model and others. Extract features, run inference, and set up retrieval in Mixpeek Studio.
Open StudioHow It Runs on Mixpeek
On Mixpeek, siglip2-so400m-patch14-384 runs as a managed extractor inside a processing pipeline. Point a bucket of zero shot image classification data at it, and Mixpeek handles GPU provisioning, batching, retries, and writing the outputs into a vector store you can query.
Extractor outputs land in the Mixpeek Vector Store (MVS), where you can combine them with retrieval, reranking, and filter stages to build end-to-end search and agent-perception pipelines, no model-serving infrastructure to maintain.
Explore on Mixpeek
Compare alternatives in this category
Hand-picked tools & platforms compared
Deep-dive technical guide
See how Mixpeek runs models as extractors
Store & search embeddings at scale
Usage-based pricing for pipelines
Compare models, APIs & infrastructure
Specification
View on HuggingFace
See model card, files, and community discussion