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 #1572] Embeddings response too slow #62898

New Issue
Closed
opened 2026-05-03 10:40:53 -05:00 by GiteaMirror · 7 comments
GiteaMirror commented 2026-05-03 10:40:53 -05:00
Owner
Copy Link

Originally created by @perezjnv on GitHub (Dec 17, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1572

I did an ingest with a CSV for fine tuning in a model called2-7b in .bin format, that worked well for me but when using ollma with a Modelfile that implements it the responses are too slow, any suggestions?

Originally created by @perezjnv on GitHub (Dec 17, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1572 I did an ingest with a CSV for fine tuning in a model called2-7b in .bin format, that worked well for me but when using ollma with a Modelfile that implements it the responses are too slow, any suggestions?
GiteaMirror added the embeddingsquestionperformance labels 2026-05-03 10:41:06 -05:00
GiteaMirror commented 2026-05-03 10:41:10 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Dec 17, 2023):

@perezjnv can you provide a sample of the CSV file and the way to do it. I will check to see if it's too slow. I think it depends on the size of the CSV, the power of your computer and the expectation of what is too slow.

<!-- gh-comment-id:1859299824 --> @igorschlum commented on GitHub (Dec 17, 2023): @perezjnv can you provide a sample of the CSV file and the way to do it. I will check to see if it's too slow. I think it depends on the size of the CSV, the power of your computer and the expectation of what is too slow.
GiteaMirror commented 2026-05-03 10:41:11 -05:00
Author
Owner
Copy Link

@perezjnv commented on GitHub (Dec 18, 2023):

Hello Igor! Thanks for answering. I am doing a conceptual test for a school system, I need to do fine tuning of documents, the largest one has 126 pages, I am doing questions and answers to ingest a model with llama2-7b.
I have carried out tests by ingesting the file llama-2-7b-chat.ggmlv3.q8_0.bin.
The idea is to have several versions of the llama2 model separately for different study subjects or other topics.
I am using a Mac m1 for development, with 16RAM

<!-- gh-comment-id:1859448393 --> @perezjnv commented on GitHub (Dec 18, 2023): Hello Igor! Thanks for answering. I am doing a conceptual test for a school system, I need to do fine tuning of documents, the largest one has 126 pages, I am doing questions and answers to ingest a model with llama2-7b. I have carried out tests by ingesting the file llama-2-7b-chat.ggmlv3.q8_0.bin. The idea is to have several versions of the llama2 model separately for different study subjects or other topics. I am using a Mac m1 for development, with 16RAM
GiteaMirror commented 2026-05-03 10:41:12 -05:00
Author
Owner
Copy Link

@easp commented on GitHub (Dec 18, 2023):

So you are using a custom, fine-tuned 7b model? What quantization level? How large are the model weights?

<!-- gh-comment-id:1861266193 --> @easp commented on GitHub (Dec 18, 2023): So you are using a custom, fine-tuned 7b model? What quantization level? How large are the model weights?
GiteaMirror commented 2026-05-03 10:41:13 -05:00
Author
Owner
Copy Link

@mchiang0610 commented on GitHub (Mar 11, 2024):

Hi @perezjnv Sorry about this. How are you ingesting the CSV file? Are you using embeddings? Ollama just started supporting embeddings, and I'm wondering if you are still hitting the same problem?

Also wondering if you are using the EMBED feature within modelfile?

<!-- gh-comment-id:1989177254 --> @mchiang0610 commented on GitHub (Mar 11, 2024): Hi @perezjnv Sorry about this. How are you ingesting the CSV file? Are you using embeddings? Ollama just started supporting embeddings, and I'm wondering if you are still hitting the same problem? Also wondering if you are using the EMBED feature within modelfile?
GiteaMirror commented 2026-05-03 10:41:14 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (May 6, 2024):

Hi there new embedding models are available now that should be much, much faster:

https://ollama.com/library/all-minilm
https://ollama.com/library/nomic-embed-text

<!-- gh-comment-id:2097103173 --> @jmorganca commented on GitHub (May 6, 2024): Hi there new embedding models are available now that should be much, much faster: https://ollama.com/library/all-minilm https://ollama.com/library/nomic-embed-text
GiteaMirror commented 2026-05-03 10:41:15 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (May 7, 2024):

Hi @perezjnv now llama3 is out. I'm also working for a school. Have you seen version 0.1.33 of Ollama, you can now launch several Ollama and several modèles at the same time.
If you could share you project on Github, it could be interesting to share with other schools.

<!-- gh-comment-id:2097607451 --> @igorschlum commented on GitHub (May 7, 2024): Hi @perezjnv now llama3 is out. I'm also working for a school. Have you seen version 0.1.33 of Ollama, you can now launch several Ollama and several modèles at the same time. If you could share you project on Github, it could be interesting to share with other schools.
GiteaMirror commented 2026-05-03 10:41:16 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Nov 30, 2024):

Hi @perezjnv, if you need help with this, I can assist as well. I'd be interested to know if you succeed.

<!-- gh-comment-id:2509401698 --> @igorschlum commented on GitHub (Nov 30, 2024): Hi @perezjnv, if you need help with this, I can assist as well. I'd be interested to know if you succeed.
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 embeddings performance question
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#62898
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?