Gemini Embedding 2
Google's first model that converts text and media into numerical form so apps can search and compare content, for developers.
This came out before trckr began following Gemini API, so it was added on Nov 8 without an alert. Its release date comes from Gemini API changelog.
Key facts
- Released
- Mar 10, 2026
- What it does
- Text and chat: Writes, answers questions and holds a conversation.Sees images: Understands pictures, screenshots and charts you share with it.Reads files: Reads documents such as PDFs.Listens: Understands speech and other sounds.Watches video: Understands video clips you share with it.
View moreView less
- Knowledge cutoff
- November 2025 · the latest events it learned about
Can I use it?
In an app
No app mentioned so far.
For developers
Developers can use it now through an API: How developers plug a model into their own apps and pay per use. A model that's "in the API" can be used by programmers, even if it isn't in a chat app yet..
Download
Not downloadable: it runs on the maker's own servers.
Price and limits
What developers pay to use it through the API: How developers plug a model into their own apps and pay per use. A model that's "in the API" can be used by programmers, even if it isn't in a chat app yet., per million Tokens: The small chunks of text a model reads and writes, about three-quarters of a word each. Prices and limits are counted in tokens. sent (input) or written back (output), and how much it can take in at once.
- Input
- $0.20
- per million tokens
- Output
- Free
- per million tokens
- Context window: How much a model can keep in mind at once: your question, any files you share and its own replies. Measured in tokens; 1,000 tokens is about 750 words.
- 8K tokens
- about 6,100 words
View moreView less
- Longest answer
- 1 tokens (about 0.75 words)
From the lab's own listing on models.dev, checked .
The details
Gemini Embedding model: A model that turns text or images into lists of numbers so software can find similar things. It powers search and recommendations; you don't chat with it. 2 Preview: An early version released for testing. It may change or disappear, and is often limited while it lasts., Google's first Multimodal: Works with more than text: a multimodal model can also look at images, listen to audio or watch video. embedding model, which turns text and other media into numbers for search and matching.
Model ID: The exact name developers type to use a model, like gpt-5-mini. New IDs often show up in a lab's systems before the official announcement, which is how trckr spots many releases early.
gemini-embedding-2-previewTimeline
/Gemini APIOfficialFirst signal
gemini-embedding-2-preview
Get alerts like this the moment they happen.
From Google and every other major lab.