NEWVectors or files. Pick a path.Start →
    Runs as a custom pluginEmbeddingVerified 2026-11-05

    OpenCLIP ViT-bigG/14 LAION-2B

    laion/CLIP-ViT-bigG-14-laion2B-39B-b160k

    OpenCLIP ViT-bigG/14 LAION-2B is a image and text embedding model (MIT). Can run on Mixpeek as a custom plugin.

    Mixpeek does not run it today. It can run as a custom plugin: the weights load without remote code.

    Record as JSON

    No demo for OpenCLIP ViT-bigG/14 LAION-2B yet

    A video, image, slideshow or playground goes here once we record one on our own data.

    What it is good and bad at

    Best at
    Largest LAION CLIP; best zero-shot accuracy in the LAION family.
    Worst at
    English-centric text tower (77 tokens); weak on text in images and counting.
    Task
    Embedding
    Input
    Image, Text
    Output
    Vector
    Use cases
    Visual search, Product search

    Templates that use it

    No template uses this model yet. Browse templates.

    Input and output

    Input

    image
    resolution
    224x224
    color
    RGB
    normalize
    CLIP mean (0.4815, 0.4578, 0.4082), std (0.2686, 0.2613, 0.2758)
    text
    max_tokens
    77
    prompts
    'a photo of a {label}.'; open_clip tokenizer, 77 tokens

    Output

    vector
    dims
    1280
    distance
    cosine
    type
    dense
    note
    image and text vectors share one space; L2-normalize before cosine

    How Mixpeek wires it

    Extractors
    None today
    Feature URIs
    None today
    Vector indexes
    None
    Retriever stages
    feature_search, mmr
    Combines with
    • video frames (VideoFrameExpander) -> image vectors -> feature_search with a text query
    • detector boxes (RT-DETR, D-FINE) -> crops -> image vectors per object
    • image vectors + E5 text vectors on captions -> feature_search fusion (dbsf)

    How it runs

    Status
    Runs as a custom plugin Not served. Would run as a custom extractor plugin (engine/extractors/<name>/v1/ with realtime.py for query embedding), loading open_clip_model.safetensors via open_clip hf-hub with trust_remote_code false. GPU recommended at this size.
    Serve apps
    None
    Runtime
    open_clip create_model_and_transforms('hf-hub:<id>')
    Weights
    safetensors, no remote code
    Parameters
    2,540 M
    Size on disk
    10,158 MB
    Preprocessing
    Resize and center-crop to 224, CLIP mean/std normalize; tokenize to 77.
    Postprocessing
    L2-normalize; cosine similarity.
    Serving
    both, batch 32
    Hardware
    Runs on CPU (slower without a GPU)

    Scale to zero

    Minimum replicas
    0
    Maximum replicas
    1
    Idle timeout
    600 s
    Memory
    16 GB
    CPUs
    2
    GPUs
    1
    Cold start, node warm
    Not recorded
    Cold start, new node
    Not recorded

    Measured cost

    Unmeasured

    No committed measurement; plan row 3 measures it in the standalone image.

    Our evaluations

    Unmeasured

    We have not evaluated this model on our own data yet. Publisher results are in the sources below.

    Questions

    What is OpenCLIP ViT-bigG/14 LAION-2B best at?
    Largest LAION CLIP; best zero-shot accuracy in the LAION family.
    What are the limits of OpenCLIP ViT-bigG/14 LAION-2B?
    English-centric text tower (77 tokens); weak on text in images and counting.
    Can OpenCLIP ViT-bigG/14 LAION-2B be used commercially?
    Yes. MIT weights; LAION notes the datasets are for research, the model license itself has no use restriction.
    How does OpenCLIP ViT-bigG/14 LAION-2B run on Mixpeek?
    Not served. Would run as a custom extractor plugin (engine/extractors/<name>/v1/ with realtime.py for query embedding), loading open_clip_model.safetensors via open_clip hf-hub with trust_remote_code false. GPU recommended at this size.
    What does OpenCLIP ViT-bigG/14 LAION-2B output?
    A 1280-dimensional dense vector, compared by cosine similarity.