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 #1269] Quantizated models availability in the default library #650

New Issue
Closed
opened 2026-04-12 10:20:35 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 10:20:35 -05:00
Owner
Copy Link

Originally created by @ghost on GitHub (Nov 24, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1269

I've being using text-generation-webui to load 3 bit quantized models to fit my GPU memory (a GTX 1650, 4GB dedicated memory). I usually go to the user TheBloke hugging Face page and check if there is a compatible model. It is a straightforward process, very easy to acomplish requiring just the path and the models name (mostly GGUF format). Most the times they provide the template for instruction tasks, and I practically don't have to touch a single line in any configuration file to work. I'm satisfied with the inference process, and I get stable results that range from 10 to 15 tokens / s.

About your project, the ollama, when I saw the available library, I couldn't load practically any model because anything >= 4 bit precision is too much for my GPU memory.

I'm here to ask, if it's not a difficult task, and if you trust users like TheBloke, to make these quantized models available to use by a simple "one click" install in your UI. These models are already available and free to use. I'm asking a practial UI to use them with few steps without the need to write configuration files for them to work.

Thanks!

Originally created by @ghost on GitHub (Nov 24, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1269 I've being using [text-generation-webui](https://github.com/oobabooga/text-generation-webui) to load 3 bit quantized models to fit my GPU memory (a GTX 1650, 4GB dedicated memory). I usually go to the user [TheBloke hugging Face](https://huggingface.co/TheBloke) page and check if there is a compatible model. It is a straightforward process, very easy to acomplish requiring just the path and the models name (mostly GGUF format). Most the times they provide the template for instruction tasks, and I practically don't have to touch a single line in any configuration file to work. I'm satisfied with the inference process, and I get stable results that range from 10 to 15 tokens / s. About your project, the ollama, when I saw the available library, I couldn't load practically any model because anything >= 4 bit precision is too much for my GPU memory. I'm here to ask, if it's not a difficult task, and if you trust users like TheBloke, to make these quantized models available to use by a simple "one click" install in your UI. These models are already available and free to use. I'm asking a practial UI to use them with few steps without the need to write configuration files for them to work. Thanks!
GiteaMirror commented 2026-04-12 10:20:36 -05:00
Author
Owner
Copy Link

@ghost commented on GitHub (Nov 24, 2023):

I'm very sorry, I'm writing this without testing the new updates (I tested long time ago). I see that there are some quantized models now.

Maybe I'm asking just a better integration with Hugging Face.

Anyway, thank you so much. Now I can test some models with your app!

<!-- gh-comment-id:1826064704 --> @ghost commented on GitHub (Nov 24, 2023): I'm very sorry, I'm writing this without testing the new updates (I tested long time ago). I see that there are some quantized models now. Maybe I'm asking just a better integration with Hugging Face. Anyway, thank you so much. Now I can test some models with your app!
GiteaMirror commented 2026-04-12 10:20:36 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Nov 24, 2023):

Thanks @guinanet, I really appreciate you creating an issue. Keep an eye on https://github.com/jmorganca/ollama/issues/1267!

<!-- gh-comment-id:1826073022 --> @jmorganca commented on GitHub (Nov 24, 2023): Thanks @guinanet, I really appreciate you creating an issue. Keep an eye on https://github.com/jmorganca/ollama/issues/1267!
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#650
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?