github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 08:59:13 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #12191] embeddinggemma doesn't have embedding tag on ollama.com #70166

New Issue
Closed
opened 2026-05-04 20:35:21 -05:00 by GiteaMirror · 9 comments
GiteaMirror commented 2026-05-04 20:35:21 -05:00
Owner
Copy Link

Originally created by @rick-github on GitHub (Sep 5, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12191

Image
Originally created by @rick-github on GitHub (Sep 5, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12191 <img width="739" height="444" alt="Image" src="https://github.com/user-attachments/assets/f7b6c7e7-2ff8-4367-ba11-1201f3f9d17c" />
GiteaMirror added the modelollama.com labels 2026-05-04 20:35:21 -05:00
GiteaMirror commented 2026-05-04 20:35:25 -05:00
Author
Owner
Copy Link

@slyapustin commented on GitHub (Sep 5, 2025):

This also cause error when using recommended ollama run embeddinggemma:

ollama run embeddinggemma
pulling manifest 
pulling 0800cbac9c20: 100% ▕█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 621 MB                         
pulling 45dc10444b87: 100% ▕█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏   34 B                         
pulling 991197bf6ec9: 100% ▕█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏  342 B                         
verifying sha256 digest 
writing manifest 
success 
Error: 400 Bad Request: "embeddinggemma" does not support generate
<!-- gh-comment-id:3257986744 --> @slyapustin commented on GitHub (Sep 5, 2025): This also cause error when using recommended `ollama run embeddinggemma`: ``` ollama run embeddinggemma pulling manifest pulling 0800cbac9c20: 100% ▕█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 621 MB pulling 45dc10444b87: 100% ▕█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 34 B pulling 991197bf6ec9: 100% ▕█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ 342 B verifying sha256 digest writing manifest success Error: 400 Bad Request: "embeddinggemma" does not support generate ```
GiteaMirror commented 2026-05-04 20:35:26 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 5, 2025):

The recommendation is incorrect because of the missing tag. Embedding models usually display ollama pull <model>.

<!-- gh-comment-id:3257994403 --> @rick-github commented on GitHub (Sep 5, 2025): The recommendation is incorrect because of the missing tag. Embedding models usually display `ollama pull <model>`.
GiteaMirror commented 2026-05-04 20:35:29 -05:00
Author
Owner
Copy Link

@slyapustin commented on GitHub (Sep 5, 2025):

Yes, and it also should not be listed in the model list in the Ollama chat:

Image
<!-- gh-comment-id:3258017395 --> @slyapustin commented on GitHub (Sep 5, 2025): Yes, and it also should not be listed in the model list in the Ollama chat: <img width="828" height="568" alt="Image" src="https://github.com/user-attachments/assets/0a78be95-d6a7-43f2-85bd-5aba75b4b984" />
GiteaMirror commented 2026-05-04 20:35:32 -05:00
Author
Owner
Copy Link

@stronk7 commented on GitHub (Sep 7, 2025):

Can this be causing the "weird" results that I'm getting when using "embeddinggemma" for embeddings in a semantic router?

Tested with many other embeddings (nomic-embed-text, bge-m3, all-minilm, ...) I get "correct" (expected) results, with very subtle differences, but with embeddinggemma I'm getting completely non-sense results. For example, for the "test" (or anything else) phrase, I get matching nodes which utterances have nothing to do with "test". Again, with practically any "phrase".

So, I bet this makes the results not really reliable, not only for my semantic router, but to any potential use of the model (RAG and friends).

Ciao :-)

<!-- gh-comment-id:3263698892 --> @stronk7 commented on GitHub (Sep 7, 2025): Can this be causing the "weird" results that I'm getting when using "embeddinggemma" for embeddings in a semantic router? Tested with many other embeddings (nomic-embed-text, bge-m3, all-minilm, ...) I get "correct" (expected) results, with very subtle differences, but with embeddinggemma I'm getting completely non-sense results. For example, for the "test" (or anything else) phrase, I get matching nodes which utterances have nothing to do with "test". Again, with practically any "phrase". So, I bet this makes the results not really reliable, not only for my semantic router, but to any potential use of the model (RAG and friends). Ciao :-)
GiteaMirror commented 2026-05-04 20:35:33 -05:00
Author
Owner
Copy Link

@uri commented on GitHub (Sep 7, 2025):

Is this also a problem?

$ curl 'http://localhost:11434/api/embed' -d '{
  "model": "embeddinggemma:latest",
  "input": "test"
}'
{"error":"this model does not support embeddings"}
<!-- gh-comment-id:3264088441 --> @uri commented on GitHub (Sep 7, 2025): Is this also a problem? ```bash $ curl 'http://localhost:11434/api/embed' -d '{ "model": "embeddinggemma:latest", "input": "test" }' {"error":"this model does not support embeddings"} ```
GiteaMirror commented 2026-05-04 20:35:33 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 7, 2025):

@uri Upgrade ollama.

<!-- gh-comment-id:3264089009 --> @rick-github commented on GitHub (Sep 7, 2025): @uri Upgrade ollama.
GiteaMirror commented 2026-05-04 20:35:33 -05:00
Author
Owner
Copy Link

@erbanku commented on GitHub (Sep 8, 2025):

@slyapustin It seems the new Ollama versions do not support running directly, or via GUI; you need to use them via an API request.
Windows PowerShell try (Needs ollama version v0.11.10 or later):

Invoke-RestMethod -Uri 'http://localhost:11434/api/embed' -Method Post -Body ('{"model":"embeddinggemma:latest","input":"test"}') -ContentType 'application/json'
Result Image
<!-- gh-comment-id:3264394727 --> @erbanku commented on GitHub (Sep 8, 2025): @slyapustin It seems the new Ollama versions do not support running directly, or via GUI; you need to use them via an API request. Windows PowerShell try (Needs ollama version v0.11.10 or later): ``` Invoke-RestMethod -Uri 'http://localhost:11434/api/embed' -Method Post -Body ('{"model":"embeddinggemma:latest","input":"test"}') -ContentType 'application/json' ``` <details> <summary>Result</summary> <img width="2010" height="1014" alt="Image" src="https://github.com/user-attachments/assets/55f31bb6-3906-4371-8853-bbc7452eb797" /> </details>
GiteaMirror commented 2026-05-04 20:35:34 -05:00
Author
Owner
Copy Link

@slyapustin commented on GitHub (Sep 9, 2025):

@erbanku It should not be listed in that dropdown at all or it must be grayed out (with indications - it's embedding model).

<!-- gh-comment-id:3270769944 --> @slyapustin commented on GitHub (Sep 9, 2025): @erbanku It should not be listed in that dropdown at all or it must be grayed out (with indications - it's embedding model).
GiteaMirror commented 2026-05-04 20:35:34 -05:00
Author
Owner
Copy Link

@slyapustin commented on GitHub (Sep 9, 2025):

Anyway, they fixed model label now

Image
<!-- gh-comment-id:3270774655 --> @slyapustin commented on GitHub (Sep 9, 2025): Anyway, they fixed model label now <img width="628" height="180" alt="Image" src="https://github.com/user-attachments/assets/fdfd1513-aec2-4e3b-9b5e-a8f0f8a99c39" />
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 model ollama.com
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#70166
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?