github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 13:19:40 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #11807] Allow exact context length configuration for OpenAI‑compatible endpoint in GUI #7834

New Issue
Open
opened 2026-04-12 20:00:10 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 20:00:10 -05:00
Owner
Copy Link

Originally created by @michael-heinrich on GitHub (Aug 8, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11807

With the introduction of the GUI, the context length for the OpenAI‑compatible endpoint is tied to the context length slider in the GUI. This introduces two related limitations:

The slider only offers a small set of discrete positions, making it impossible to set an arbitrary, precise value.

Why this matters:

With the coarse steps, either my VRAM is not fully utilized, or the model spills significantly into RAM and there is no way to fine-tune the context length for my GPU. This particularly affects the utility of VS Code agent mode because it cannot be configured to use the native ollama endpoint, but requires an OpenAI compatible endpoint.

Proposed solutions:

  1. Provide a way to manually enter an exact num_ctx value in the GUI, in addition to the slider.

  2. Or, allow the slider to move freely between the snaps.

Thanks for considering my proposal.

Originally created by @michael-heinrich on GitHub (Aug 8, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11807 With the introduction of the GUI, the context length for the OpenAI‑compatible endpoint is tied to the context length slider in the GUI. This introduces two related limitations: The slider only offers a small set of discrete positions, making it impossible to set an arbitrary, precise value. Why this matters: With the coarse steps, either my VRAM is not fully utilized, or the model spills significantly into RAM and there is no way to fine-tune the context length for my GPU. This particularly affects the utility of VS Code agent mode because it cannot be configured to use the native ollama endpoint, but requires an OpenAI compatible endpoint. Proposed solutions: 1. Provide a way to manually enter an exact num_ctx value in the GUI, in addition to the slider. 2. Or, allow the slider to move freely between the snaps. Thanks for considering my proposal.
GiteaMirror added the appcontext-lengthfeature request labels 2026-04-12 20:00:10 -05:00
GiteaMirror commented 2026-04-12 20:00:11 -05:00
Author
Owner
Copy Link

@sunhy0316 commented on GitHub (Aug 8, 2025):

If I don't use the app and instead use ollama serve, will the environment variable OLLAMA_CONTEXT_LENGTH or the PARAMETER num_ctx in the Modelfile still work?

<!-- gh-comment-id:3167849446 --> @sunhy0316 commented on GitHub (Aug 8, 2025): If I don't use the app and instead use `ollama serve`, will the environment variable `OLLAMA_CONTEXT_LENGTH` or the `PARAMETER num_ctx` in the Modelfile still work?
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 app context-length feature request
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#7834
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?