github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #3749] Rerankers and Embeddings #48824

New Issue
Closed
opened 2026-04-28 09:35:38 -05:00 by GiteaMirror · 17 comments
GiteaMirror commented 2026-04-28 09:35:38 -05:00
Owner
Copy Link

Originally created by @BradKML on GitHub (Apr 19, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3749

Hello from the agentic AI crowd, other than that Ollama does not have concurrent model abilities, is it possible to get Ollama to run Rerankers (like Cohere and Jina) or text embeddings (SentenceTransformers e.g. SBERT) for applications that does not directly use LLMs?

Originally created by @BradKML on GitHub (Apr 19, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3749 Hello from the agentic AI crowd, other than that Ollama does not have concurrent model abilities, is it possible to get Ollama to run Rerankers (like Cohere and Jina) or text embeddings (SentenceTransformers e.g. SBERT) for applications that does not directly use LLMs?
GiteaMirror added the model label 2026-04-28 09:35:38 -05:00
GiteaMirror commented 2026-04-28 09:35:44 -05:00
Author
Owner
Copy Link

@thinkverse commented on GitHub (Apr 19, 2024):

I'm not sure about Rerankers but Ollama started supporting text embeddings as of 0.1.26 and even released a blog post about Embedding models. Here's a short list of some currently available models:

The community has also added some more models, searching embed gives a fair amount of results.

More will become available later hopefully once more architectures are supported.

<!-- gh-comment-id:2066994206 --> @thinkverse commented on GitHub (Apr 19, 2024): I'm not sure about Rerankers but Ollama started supporting text embeddings as of [0.1.26](https://github.com/ollama/ollama/releases/tag/v0.1.26) and even released a blog post about [Embedding models](https://ollama.com/blog/embedding-models). Here's a short list of some currently available models: - [snowflake-arctic-embed](https://ollama.com/library/snowflake-arctic-embed) - [mxbai-embed-large](https://ollama.com/library/mxbai-embed-large) - [nomic-embed-text](https://ollama.com/library/nomic-embed-text) - [all-minilm](https://ollama.com/library/all-minilm) The community has also added some more models, [searching embed](https://ollama.com/search?q=embed&p=1) gives a fair amount of results. More will become available later hopefully once more architectures are supported.
GiteaMirror commented 2026-04-28 09:35:46 -05:00
Author
Owner
Copy Link

@BradKML commented on GitHub (Apr 24, 2024):

Should've focused on Rerankers

As for previously noted issues there are these two https://github.com/ollama/ollama/issues/3630 https://github.com/ollama/ollama/issues/3631

<!-- gh-comment-id:2074230480 --> @BradKML commented on GitHub (Apr 24, 2024): Should've focused on Rerankers - LocalAI is also waiting for support for Rerankers https://github.com/mudler/LocalAI/issues/2018 - Xorbit Xinference is already https://github.com/xorbitsai/inference/pull/672 As for previously noted issues there are these two https://github.com/ollama/ollama/issues/3630 https://github.com/ollama/ollama/issues/3631
GiteaMirror commented 2026-04-28 09:35:46 -05:00
Author
Owner
Copy Link

@3SMMZRjWgS commented on GitHub (May 1, 2024):

I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end.

<!-- gh-comment-id:2087740526 --> @3SMMZRjWgS commented on GitHub (May 1, 2024): I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end.
GiteaMirror commented 2026-04-28 09:35:47 -05:00
Author
Owner
Copy Link

@uzumakinaruto19 commented on GitHub (May 8, 2024):

I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end.

+1

<!-- gh-comment-id:2100449483 --> @uzumakinaruto19 commented on GitHub (May 8, 2024): > I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end. +1
GiteaMirror commented 2026-04-28 09:35:55 -05:00
Author
Owner
Copy Link

@cope commented on GitHub (May 8, 2024):

I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end.

+1

<!-- gh-comment-id:2100634981 --> @cope commented on GitHub (May 8, 2024): > I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end. +1
GiteaMirror commented 2026-04-28 09:35:57 -05:00
Author
Owner
Copy Link

@stronk7 commented on GitHub (May 8, 2024):

Heh,

I'd suggest to stop adding comments and, instead, just upvote the proposal, it's a little bit annoying to get emails with nothing but a +1.

Thanks :-)

<!-- gh-comment-id:2100772254 --> @stronk7 commented on GitHub (May 8, 2024): Heh, I'd suggest to stop adding comments and, instead, just upvote the proposal, it's a little bit annoying to get emails with nothing but a +1. Thanks :-)
GiteaMirror commented 2026-04-28 09:36:01 -05:00
Author
Owner
Copy Link

@RLinnae commented on GitHub (May 20, 2024):

Would also be great to see rerankers as now I have other packages installed just for re-ranking. Making ollama feel like a waste.

<!-- gh-comment-id:2120800968 --> @RLinnae commented on GitHub (May 20, 2024): Would also be great to see rerankers as now I have other packages installed just for re-ranking. Making ollama feel like a waste.
GiteaMirror commented 2026-04-28 09:36:04 -05:00
Author
Owner
Copy Link

@BradKML commented on GitHub (May 21, 2024):

@RLinnae seconding this but what are the options out there atm that is not so complex that you need a Docker instance?

<!-- gh-comment-id:2121656428 --> @BradKML commented on GitHub (May 21, 2024): @RLinnae seconding this but what are the options out there atm that is not so complex that you need a Docker instance?
GiteaMirror commented 2026-04-28 09:36:11 -05:00
Author
Owner
Copy Link

@yushengliao commented on GitHub (Jun 7, 2024):

+1

<!-- gh-comment-id:2153782088 --> @yushengliao commented on GitHub (Jun 7, 2024): +1
GiteaMirror commented 2026-04-28 09:36:14 -05:00
Author
Owner
Copy Link

@RLinnae commented on GitHub (Jun 8, 2024):

@RLinnae seconding this but what are the options out there atm that is not so complex that you need a Docker instance?

I am mostly using flashrank at the moment.

<!-- gh-comment-id:2155980247 --> @RLinnae commented on GitHub (Jun 8, 2024): > @RLinnae seconding this but what are the options out there atm that is not so complex that you need a Docker instance? I am mostly using flashrank at the moment.
GiteaMirror commented 2026-04-28 09:36:17 -05:00
Author
Owner
Copy Link

@zydmtaichi commented on GitHub (Jun 28, 2024):

I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end.

+1

<!-- gh-comment-id:2195899829 --> @zydmtaichi commented on GitHub (Jun 28, 2024): > I'd like to add a vote to @BradKML's suggestion on a re-ranker feature in the Ollama framework. This will help keep any RAG pipeline within Ollama end-to-end. +1
GiteaMirror commented 2026-04-28 09:36:19 -05:00
Author
Owner
Copy Link

@BradKML commented on GitHub (Jul 4, 2024):

Can people please stop with the "+1" already and suggest models that should be included instead? (with reference link to corresponding resources)

<!-- gh-comment-id:2207840950 --> @BradKML commented on GitHub (Jul 4, 2024): Can people please stop with the "+1" already and suggest models that should be included instead? (with reference link to corresponding resources)
GiteaMirror commented 2026-04-28 09:36:20 -05:00
Author
Owner
Copy Link

@mavwolverine commented on GitHub (Jul 10, 2024):

From https://docs.continue.dev/setup/configuration#best-overall-experience:

https://huggingface.co/voyageai/rerank-1/tree/main
https://huggingface.co/voyageai/voyage-code-2

<!-- gh-comment-id:2221579962 --> @mavwolverine commented on GitHub (Jul 10, 2024): From https://docs.continue.dev/setup/configuration#best-overall-experience: https://huggingface.co/voyageai/rerank-1/tree/main https://huggingface.co/voyageai/voyage-code-2
GiteaMirror commented 2026-04-28 09:36:22 -05:00
Author
Owner
Copy Link

@tikendraw commented on GitHub (Aug 21, 2024):

add reranker models please.

<!-- gh-comment-id:2302565264 --> @tikendraw commented on GitHub (Aug 21, 2024): add reranker models please.
GiteaMirror commented 2026-04-28 09:36:24 -05:00
Author
Owner
Copy Link

@Nordln commented on GitHub (Aug 26, 2024):

+1 for adding reranker models to Ollama. Same issue was raised here: https://github.com/ollama/ollama/issues/4510 and here: https://github.com/ggerganov/llama.cpp/issues/8555

<!-- gh-comment-id:2309837246 --> @Nordln commented on GitHub (Aug 26, 2024): +1 for adding reranker models to Ollama. Same issue was raised here: https://github.com/ollama/ollama/issues/4510 and here: https://github.com/ggerganov/llama.cpp/issues/8555
GiteaMirror commented 2026-04-28 09:36:26 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Sep 2, 2024):

Merging with https://github.com/ollama/ollama/issues/3368

Thanks for the issue and comments all!

<!-- gh-comment-id:2325295688 --> @jmorganca commented on GitHub (Sep 2, 2024): Merging with https://github.com/ollama/ollama/issues/3368 Thanks for the issue and comments all!
GiteaMirror commented 2026-04-28 09:36:28 -05:00
Author
Owner
Copy Link

@sandstad commented on GitHub (Jan 11, 2025):

Working wit a RAGFlow project and would have been great to get re-ranking support in Ollama

<!-- gh-comment-id:2585307558 --> @sandstad commented on GitHub (Jan 11, 2025): Working wit a RAGFlow project and would have been great to get re-ranking support in Ollama
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
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#48824
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?