Gemini Embedding 2 is Google's embedding model with a 8K context window and up to 128 output tokens, available from 3 providers, starting at $0.200 / 1M input. Google's first fully multimodal embedding model capable of mapping text, images, video, audio, and PDFs into a unified vector space.
Specifications
Canonical IDgoogle-gemini-2-embedding
TypeEmbedding
StatusActive
CreatorGoogleGoogle
Providers
Context Window8K tokens
Max Output128 tokens
Input ModalitiesAudioImagePdfTextVideo
Output ModalitiesEmbedding
Embedding Dimensions3072
Release Date · 3 months ago
Knowledge Cutoff

Capabilities

Input5/5
Text
Image
Audio
Video
PDF
Output1/5
Text·
Image·
Audio·
Video·
Embedding
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Pricing by Provider

ProviderStandardBatch
Input
$ / 1M
Audio In
$ / 1M
Input
$ / 1M
Audio In
$ / 1M
Google Gemini logo
Google Gemini
gemini-embedding-2
$0.200$6.50$0.100$3.25
Google Vertex AI logo
Google Vertex AI
gemini-embedding-2
$0.200$6.50$0.100$3.25
Vercel AI Gateway logo
Vercel AI Gateway
google/gemini-embedding-2
$0.200N/A

Cost Calculator

Preset:

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Text Embedding 52K$0.025Available
Embed 4128K$0.120$0.470Available
Embed 4 Img$0.470Available
Embed 4 Txt$0.120Available
Text Embedding 42K$0.100Deprecated
Voyage 432K$0.060Available
Voyage 4 Large32K$0.120Available
Voyage 4 Lite32K$0.020Available
Voyage 3.532K$0.060Available
Voyage 3.5 Lite32K$0.020Available
Gemini Embedding 28K$0.200Current

Model IDs