NEWVectors or files. Pick a path.Start →
    Models/Zero Shot Image Classification/google/siglip2-base-patch16-224
    Zero Shot Image Classificationtransformersapache-2.0

    siglip2-base-patch16-224

    by google

    SigLIP 2 at base size and fixed 224px, the cheap end of the family

    Identifier
    Model ID
    google/siglip2-base-patch16-224

    Overview

    SigLIP 2 replaces CLIP's softmax contrastive loss with a pairwise sigmoid loss, which removes the need for a global normalization across the batch and trains stably at scale. This is the base checkpoint at fixed 224x224 input.

    Against the NaFlex variants it gives up native aspect ratio handling, which matters most on documents, screenshots and wide video frames where squashing to a square distorts the spatial relationships the encoder is trying to capture. What it gains is cost: at 375M parameters it is the checkpoint you can afford to run over every frame rather than a keyframe.

    Apache 2.0.

    Architecture

    SigLIP 2 base, 16x16 patches, fixed 224x224 input, 375,187,970 parameters. Zero-shot-image-classification pipeline with an aligned image-text embedding space.

    Key Capabilities

    • Cross-modal retrieval: text query against an image corpus
    • Zero-shot classification against a label list that changes without retraining
    • Image-to-image similarity through the vision tower
    • Cheap enough for per-frame video indexing

    Use Cases on Mixpeek

    • Text-to-image search over a large library on a fixed budget
    • Per-frame video indexing where a larger encoder is too slow
    • Zero-shot tagging against a category list that changes often
    • First-stage recall ahead of a larger encoder for reranking

    Tags

    transformerssafetensorssiglipvisionzero-shot-image-classificationarxiv:2502.14786arxiv:2303.15343arxiv:2209.06794license:apache-2.0endpoints_compatibleregion:us

    Use siglip2-base-patch16-224 on Mixpeek

    Build multimodal processing pipelines with this model and others. Extract features, run inference, and set up retrieval in Mixpeek Studio.

    Open Studio

    How It Runs on Mixpeek

    On Mixpeek, siglip2-base-patch16-224 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.