github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #2915] All Models running slow on Ollama #63826

New Issue
Closed
opened 2026-05-03 15:05:31 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-05-03 15:05:31 -05:00
Owner
Copy Link

Originally created by @RahulVerma989 on GitHub (Mar 4, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2915

Hi there,

Till now I have tested all these models on my VPS machine and all of them are running extremely slow, taking minute on an average to get a full response.

Screenshot 2024-03-04 at 8 46 02 PM Screenshot 2024-03-04 at 8 51 26 PM

My VPS Specs:
RAM: 32 GB
CPU: 8
Bandwidth: 33 TB
Disk space: 400 GB

Is there any model which will provide fast responses on ollama?

Originally created by @RahulVerma989 on GitHub (Mar 4, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2915 Hi there, Till now I have tested all these models on my VPS machine and all of them are running extremely slow, taking minute on an average to get a full response. <img width="453" alt="Screenshot 2024-03-04 at 8 46 02 PM" src="https://github.com/ollama/ollama/assets/48805821/39569bff-2e2a-4dc6-8c72-32ba49b45a25"> <img width="850" alt="Screenshot 2024-03-04 at 8 51 26 PM" src="https://github.com/ollama/ollama/assets/48805821/23e07b72-6fac-4640-8092-0cca15fbe94f"> My VPS Specs: RAM: 32 GB CPU: 8 Bandwidth: 33 TB Disk space: 400 GB Is there any model which will provide fast responses on ollama?
GiteaMirror commented 2026-05-03 15:05:34 -05:00
Author
Owner
Copy Link

@RahulVerma989 commented on GitHub (Mar 4, 2024):

Phi:chat is running a little better than other models.

Screenshot 2024-03-05 at 12 21 29 AM Screenshot 2024-03-05 at 12 23 58 AM
<!-- gh-comment-id:1977249053 --> @RahulVerma989 commented on GitHub (Mar 4, 2024): Phi:chat is running a little better than other models. <img width="1392" alt="Screenshot 2024-03-05 at 12 21 29 AM" src="https://github.com/ollama/ollama/assets/48805821/7b75023d-d7ab-4f41-a640-9910444fdbf6"> <img width="738" alt="Screenshot 2024-03-05 at 12 23 58 AM" src="https://github.com/ollama/ollama/assets/48805821/b4a8e289-b0b0-4c2b-b81c-64cdc9973b8f">
GiteaMirror commented 2026-05-03 15:05:36 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 4, 2024):

Hey @RahulVerma989 . You'll get a lot better response times using a system w/ a GPU such as an AWS EC2 P4/G4 instance. You can also try to use a higher quantization level such as gemma:7b-chat-q2_K, although that might not give you great results. As you mentioned, a lower parameter count model such as phi will also work.

Those are essentially the three levers you've got though:

  1. better hardware (such as a GPU)
  2. higher quantization (i.e. "compression") level (q2_K, q3_K_S, q3_K_M, q3_K_L)
  3. lower parameter count models

I'm going to go ahead and close the issue. Feel free to keep commenting though, or ask away on the Discord.

<!-- gh-comment-id:1977472980 --> @pdevine commented on GitHub (Mar 4, 2024): Hey @RahulVerma989 . You'll get a lot better response times using a system w/ a GPU such as an AWS EC2 P4/G4 instance. You can also try to use a higher quantization level such as `gemma:7b-chat-q2_K`, although that might not give you great results. As you mentioned, a lower parameter count model such as `phi` will also work. Those are essentially the three levers you've got though: 1. better hardware (such as a GPU) 2. higher quantization (i.e. "compression") level (q2_K, q3_K_S, q3_K_M, q3_K_L) 3. lower parameter count models I'm going to go ahead and close the issue. Feel free to keep commenting though, or ask away on the Discord.
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
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#63826
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?