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 #11124] Ollama is using Too high resource with ChromaDB when vector_store.similarity_search_with_score is used. #33096

New Issue
Closed
opened 2026-04-22 15:22:59 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-22 15:22:59 -05:00
Owner
Copy Link

Originally created by @doyoungim999 on GitHub (Jun 18, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11124

What is the issue?

I am using chroma with OllamaEmbeddings for similarity_search_with_score.
I found Chroma and Ollama is using High CPU ( 20cores of 32 Cores) . Is this common?
My text is 5K bytes only.

Is there any alternative to use less CPU ?
I am using embedding_fution of Chroma. Ollama process uses too high CPU like 23 cores of 32 cores. This is too high. Is this normal with model llama3.2? Do I need to change model for embeddings?

embeddings = OllamaEmbeddings(
model='llama3.2',
)
vector_store = Chroma(embedding_function=embeddings, persist_directory=vectorstore_path, collection_name=collection_name)

def vector_search(query):

CONTEXT=""
results = vector_store.similarity_search_with_score(
query, k=4
)
CONTEXT = results
return CONTEXT

Relevant log output

a lot of CPU usage like 20core with Ollama.
ollama process used 23 core of 32 Cores.

OS

OpenShift with CoreOS

GPU

No.

CPU

No response

Ollama version

ollama version is 0.6.2

Originally created by @doyoungim999 on GitHub (Jun 18, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11124 ### What is the issue? I am using chroma with OllamaEmbeddings for similarity_search_with_score. I found Chroma and Ollama is using High CPU ( 20cores of 32 Cores) . Is this common? My text is 5K bytes only. Is there any alternative to use less CPU ? I am using embedding_fution of Chroma. Ollama process uses too high CPU like 23 cores of 32 cores. This is too high. Is this normal with model llama3.2? Do I need to change model for embeddings? embeddings = OllamaEmbeddings( model='llama3.2', ) vector_store = Chroma(embedding_function=embeddings, persist_directory=vectorstore_path, collection_name=collection_name) def vector_search(query): CONTEXT="" results = vector_store.similarity_search_with_score( query, k=4 ) CONTEXT = results return CONTEXT ### Relevant log output ```shell a lot of CPU usage like 20core with Ollama. ollama process used 23 core of 32 Cores. ``` ### OS OpenShift with CoreOS ### GPU No. ### CPU _No response_ ### Ollama version ollama version is 0.6.2
GiteaMirror added the bug label 2026-04-22 15:22:59 -05:00
GiteaMirror commented 2026-04-22 15:23:03 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 18, 2025):

Try using an embedding model for the embeddings. You can use any model for generating embeddings. but models dedicated to embeddings require less resources.

<!-- gh-comment-id:2986018860 --> @rick-github commented on GitHub (Jun 18, 2025): Try using an [embedding model](https://ollama.com/search?c=embedding) for the embeddings. You can use any model for generating embeddings. but models dedicated to embeddings require less resources.
GiteaMirror commented 2026-04-22 15:23:03 -05:00
Author
Owner
Copy Link

@doyoungim999 commented on GitHub (Jun 23, 2025):

Hi
I tested llama3.2 model and bge-m3 model for my data retrieving from chorma DB.

  • llama 3.2 is using almost 30% of my ubuntu CPU.
  • bge-m3 is using 5% of my ubuntu CPU.

What makes this big difference of CPU usage ?
bge-m3 does not support generate vs llama3.2 support generate and embeddings?

<!-- gh-comment-id:2995229376 --> @doyoungim999 commented on GitHub (Jun 23, 2025): Hi I tested llama3.2 model and bge-m3 model for my data retrieving from chorma DB. - llama 3.2 is using almost 30% of my ubuntu CPU. - bge-m3 is using 5% of my ubuntu CPU. What makes this big difference of CPU usage ? bge-m3 does not support generate vs llama3.2 support generate and embeddings?
GiteaMirror commented 2026-04-22 15:23:04 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 23, 2025):

models dedicated to embeddings require less resources.

<!-- gh-comment-id:2995245873 --> @rick-github commented on GitHub (Jun 23, 2025): > models dedicated to embeddings require less resources.
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 bug
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#33096
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?