Tools · Gemini AI Tools

Multimodal Similarity

☁ Cloud AI

Compare two items when the configured embedding model supports them. gemini-embedding-2 accepts text and text documents. Image, audio, and video combinations fail closed.

Powered by Google Gemini. DevNestro is not an official Google product.

This reflects embedding similarity and is not proof that two texts have the same meaning. Model: gemini-embedding-2

How it works

  1. Choose text or a text document for each side. Image, audio, and video stay hidden from ranking unless the embedding model supports them.
  2. Confirm cloud embeddings. Unsupported combinations fail closed without pretending a score exists.
  3. Read the cosine score as a vector distance, not a meaning verdict.

FAQ

Is this processed locally?

No. Supported items are sent to Google Gemini for embeddings. Ranking is not processed only in your browser.

Why did an image comparison fail?

The configured model is gemini-embedding-2, which DevNestro treats as text embeddings. Image, audio, and video combinations fail closed.

Read the original guide: Compare supported items only, and fail closed when the embedding model cannot.

An unhandled error has occurred. Reload Dismiss

Rejoining the server...

Rejoin failed... trying again in seconds.

Failed to rejoin.
Please retry or reload the page.

The session has been paused by the server.

Failed to resume the session.
Please retry or reload the page.