github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7677] Enable image embeddings for vision models #51409

New Issue
Closed
opened 2026-04-28 19:54:41 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-28 19:54:41 -05:00
Owner
Copy Link

Originally created by @kevin-pw on GitHub (Nov 15, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7677

I would love to be able to create embeddings for images with vision models like llama3.2-vision.

Creating image and text embeddings with a vision-capable model should allow creating image search and image categorization applications.

If my understanding of the shared semantic vector space of image models is correct, it should be possible to perform calculations like cosine similarity on text and image embeddings to, for example, find all the photos of puppy dogs in a random assortment of photos :)

At this time, the generate endpoint accepts an images parameter, but the embed endpoint does not. I tried passing an image as a base64 string to the input parameter of the embed endpoint, but the resulting embedding appears to be the vector of the text string and not of the image.

Would it be possible to expand the embed endpoint to accept an image parameter?

Originally created by @kevin-pw on GitHub (Nov 15, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7677 I would love to be able to create embeddings for images with vision models like `llama3.2-vision`. Creating image and text embeddings with a vision-capable model should allow creating image search and image categorization applications. If my understanding of the shared semantic vector space of image models is correct, it should be possible to perform calculations like cosine similarity on text and image embeddings to, for example, find all the photos of puppy dogs in a random assortment of photos :) At this time, the `generate` endpoint accepts an `images` parameter, but the `embed` endpoint does not. I tried passing an image as a base64 string to the `input` parameter of the `embed` endpoint, but the resulting embedding appears to be the vector of the text string and not of the image. Would it be possible to expand the `embed` endpoint to accept an image parameter?
GiteaMirror added the feature request label 2026-04-28 19:54:41 -05:00
GiteaMirror commented 2026-04-28 19:55:27 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 15, 2024):

https://github.com/ollama/ollama/issues/5304

<!-- gh-comment-id:2478352842 --> @rick-github commented on GitHub (Nov 15, 2024): https://github.com/ollama/ollama/issues/5304
GiteaMirror commented 2026-04-28 19:55:29 -05:00
Author
Owner
Copy Link

@kevin-pw commented on GitHub (Nov 15, 2024):

Thank you for pointing me to the existing issue on my feature request. I am closing this issue as duplicate.

<!-- gh-comment-id:2479461785 --> @kevin-pw commented on GitHub (Nov 15, 2024): Thank you for pointing me to the existing issue on my feature request. I am closing this issue as duplicate.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#51409
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?