1.  [](/)
2.  [Models](/models)
3.  VideoPrism Base (16 frames, 288px)

Not runnable todayEmbeddingVerified 2026-11-05

# VideoPrism Base (16 frames, 288px)

google/videoprism-base-f16r288

VideoPrism Base (16 frames, 288px) is a video embedding model (Apache-2.0). Not runnable on Mixpeek today.

Something blocks it today, such as the license, remote code or a GPU-only runtime. The reason is below.

[Record as JSON](/models/videoprism-base-f16r288.json)

No demo for VideoPrism Base (16 frames, 288px) 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

General-purpose frozen video encoder: strong features for classification, localization and retrieval probes.

Worst at

JAX/Flax only; no text tower, so no text queries: video-to-video similarity or trained probes only.

Task

Embedding

Input

Video

Output

Vector

Use cases

Video understanding, Action recognition

## Templates that use it

No template uses this model yet. [Browse templates](/templates).

## Input and output

### Input

Video

Frames

16

Resolution

288x288

### Output

Vector

Dims

per model config (not verified here)

Distance

cosine

Type

dense

## How Mixpeek wires it

Extractors

None today

Feature URIs

None today

Vector indexes

None

Retriever stages

None

Combines with

*   scene splits -> VideoPrism clip vectors -> feature\_search by example clip

## How it runs

Status

Not runnable today Weights are Flax .npz arrays loaded by the videoprism JAX package; Mixpeek's runtime is PyTorch, so it needs JAX in the image or a port. Weights-only arrays, Apache-2.0, so no license block.

Serve apps

None

Runtime

videoprism (JAX/Flax)

Weights

JAX, no remote code

Parameters

Not recorded

Size on disk

458 MB

Preprocessing

Sample 16 frames, resize to 288, scale to 0..1.

Postprocessing

Global pooled embedding; L2-normalize for retrieval.

Serving

batch

Hardware

Runs on CPU (slower without a GPU)

### Scale to zero

Minimum replicas

0

Maximum replicas

1

Idle timeout

600 s

Memory

6 GB

CPUs

2

GPUs

0

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 VideoPrism Base (16 frames, 288px) best at?

General-purpose frozen video encoder: strong features for classification, localization and retrieval probes.

What are the limits of VideoPrism Base (16 frames, 288px)?

JAX/Flax only; no text tower, so no text queries: video-to-video similarity or trained probes only.

Can VideoPrism Base (16 frames, 288px) be used commercially?

Yes. Apache-2.0 per the hub tag and repo.

How does VideoPrism Base (16 frames, 288px) run on Mixpeek?

Weights are Flax .npz arrays loaded by the videoprism JAX package; Mixpeek's runtime is PyTorch, so it needs JAX in the image or a port. Weights-only arrays, Apache-2.0, so no license block.

## Identity

Catalog id

`videoprism-base-f16r288`

Family

videoprism

Hub id

`google/videoprism-base-f16r288`

Variants

google/videoprism-large-f8r288

License

Apache-2.0

Commercial use

AllowedApache-2.0 per the hub tag and repo.

Released

2025-06

Last verified

Verified 2026-11-05

Paper

[arxiv.org/abs/2402.13217](https://arxiv.org/abs/2402.13217)

Code

[github.com/google-deepmind/videoprism](https://github.com/google-deepmind/videoprism)

## References and formats

*   `hf://google/videoprism-base-f16r288@be719a406d563b66f0ac969e7c94bab8e997c81a`

Format

File

Loads today

JAX

flax\_base\_f16r288\_repeated.npz

No

## Related models

*   [VideoPrism LvT Base (video-text, 16 frames, 288px)](/models/videoprism-lvt-base-f16r288)Embedding
*   [Multilingual E5 Large Instruct](/models/multilingual-e5-large-instruct)Embedding
*   [Gemini Embedding 2](/models/gemini-embedding-2)Embedding
*   [all-MiniLM-L6-v2](/models/all-minilm-l6-v2)Embedding
*   [Vertex AI multimodalembedding@001](/models/multimodalembedding)Embedding
*   [SigLIP Base patch16 224](/models/siglip-base-patch16-224)Embedding

## Sources

*   [huggingface.co/google/videoprism-base-f16r288](https://huggingface.co/google/videoprism-base-f16r288)
*   [arxiv.org/abs/2402.13217](https://arxiv.org/abs/2402.13217)
*   [github.com/google-deepmind/videoprism](https://github.com/google-deepmind/videoprism)