Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Google / Google Embeddings

Multimodal Embedding 001

Maps text, images and video segments into embeddings for cross-modal search; video audio is not processed.

deprecatedClosed weightsEmbeddingsSemantic search
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
14 of 26 specification fields collected
Input token limit3232 tokens
Vector dimensionsText/image: 128, 256, 512 or 1,408; video: 1,408See specification notes below
Input billingSee ratesGoogle Enterprise Agent Platform (online)
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Input token limit
Not collected

Media inputs have separate limits; no unified token budget is established.

Maximum output
Not collected
Input modalities
Text, Image, VideoSource for Input modalities (opens in a new tab)
Context caching
Not collected
Reasoning
Not collected
Tool calling
Not collected
Structured output
Not collected
Streaming
Not collected
Batch processing
Not collected
Fine-tuning
Not collected
Knowledge cutoff
Not collected
Parameters
Not collected
Architecture
Not collected
Maximum input tokens
32 tokensSource for Maximum input tokens (opens in a new tab)

Text component only; this is not a 32-token budget for image/video inputs.

Embedding dimensions
Text/image: 128, 256, 512 or 1,408; video: 1,408Source for Embedding dimensions (opens in a new tab)
Embedding controls
Set parameters.dimension for reduced text/image vectors. Longer text is internally shortened to 32 tokens.Source for Embedding controls (opens in a new tab)
Embedding compatibility
Text and image vectors occupy a shared semantic space for cross-modal retrieval.Source for Embedding compatibility (opens in a new tab)
Image input methods
Base64, Cloud Storage URISource for Image input methods (opens in a new tab)
Image input limits
BMP, GIF, JPG or PNG. 20 MB maximum; base64 limit applies after PNG transcoding, Cloud Storage limit to the original file. Images are resized to 512 × 512.Source for Image input limits (opens in a new tab)
Source video limits
Cloud Storage videos may be longer, but only 120 seconds can be analyzed at once. Supported containers: AVI, FLV, MKV, MOV, MP4, MPEG, MPG, WEBM, WMV. Audio content is ignored.Source for Source video limits (opens in a new tab)
Embedding output modes
Video Essential: intervals ≥15 seconds, up to 4 vectors/minute. Standard: 8 to <15 seconds, up to 8/minute. Plus: 4 to <8 seconds, up to 15/minute.Source for Embedding output modes (opens in a new tab)
Request limits
Regional default quota is 120–600 requests/minute/project; quota increases may be requested. Video interval minimum is 4 seconds.Source for Request limits (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Google Enterprise Agent Platform (online)

Provider source (opens in a new tab)
API model ID
multimodalembedding@001
Access
Google Cloud project with billing and Agent Platform API enabled; IAM and regional quotas apply. Online prediction.
Regions
Region availability and project quotas must be checked for this exact model
audioUnderstanding
Not supportedSource for audioUnderstanding (opens in a new tab)

Video audio is ignored.

videoWindow
120 seconds per analysisSource for videoWindow (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Embedding pricing: billing unit needs verificationText input1$0.00021,000 provider billing counts (unit unresolved)
Embedding pricing: billing unit needs verificationImage input1$0.0001provider billing count (unit unresolved)
Embedding pricing: billing unit needs verificationVideo Plus1$0.002provider billing count (unit unresolved)
Embedding pricing: billing unit needs verificationVideo Standard1$0.001provider billing count (unit unresolved)
Embedding pricing: billing unit needs verificationVideo Essential1$0.0005provider billing count (unit unresolved)

Rate conditions

  1. The current table prints count rather than a modality-specific unit. Earlier prose describes image/second/character billing, but exact current SKU denominators require verification; no token conversion is made.
Pricing source (opens in a new tab)

Data & enterprise terms

Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Google Enterprise Agent Platform (online)

multimodalembedding@001
Not verified

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

How it works
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

Minimum reusable input
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

Lifetime & refresh
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

How to enable it
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

Usage fields
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

Billing rules
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

Limits & data boundaries
Not collected

Caching for this embedding endpoint is not established by the reviewed model docs. Generative Gemini prompt-cache support is not assumed.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 14 Oct 2026

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Google identifies this proprietary, Google-hosted model separately from open embedding models. No downloadable weights or open-weight license is supplied in the reviewed documentation; closed distribution is inferred, not a claim about unpublished architecture.

Classification evidence (opens in a new tab)

Lifecycle dates

Enterprise general availability12 Feb 2024Source (opens in a new tab)
Scheduled Enterprise retirement1 Apr 2027Source (opens in a new tab)

Sources reviewed