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 #2722] How can I specify the context window size using OpenAI compatible API? #1633

New Issue
Closed
opened 2026-04-12 11:34:17 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 11:34:17 -05:00
Owner
Copy Link

Originally created by @egoist on GitHub (Feb 24, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2722

I wonder if there's a way to apply context size to https://github.com/ollama/ollama/blob/main/docs/openai.md

Originally created by @egoist on GitHub (Feb 24, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2722 I wonder if there's a way to apply context size to https://github.com/ollama/ollama/blob/main/docs/openai.md
GiteaMirror added the question label 2026-04-12 11:34:17 -05:00
GiteaMirror commented 2026-04-12 11:34:18 -05:00
Author
Owner
Copy Link

@chigkim commented on GitHub (Jun 26, 2024):

Did you find an answer?
I thought I might get luck with max_tokens parameter, but it doesn't seem to do anything according to log.
It doesn't set --num-predict for llama.cpp either.

<!-- gh-comment-id:2191587435 --> @chigkim commented on GitHub (Jun 26, 2024): Did you find an answer? I thought I might get luck with max_tokens parameter, but it doesn't seem to do anything according to log. It doesn't set --num-predict for llama.cpp either.
GiteaMirror commented 2026-04-12 11:34:18 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Jul 18, 2024):

Hey guys, sorry for the slow response. You should be able to use the max_tokens field to control the context size.

<!-- gh-comment-id:2237739427 --> @pdevine commented on GitHub (Jul 18, 2024): Hey guys, sorry for the slow response. You should be able to use the `max_tokens` field to control the context size.
GiteaMirror commented 2026-04-12 11:34:19 -05:00
Author
Owner
Copy Link

@chigkim commented on GitHub (Jul 19, 2024):

If max_tokens controls the context size, how can we control num_predict with OpenAI API?
Isn't max_tokens on OpenAI API supposed to control how many tokens to generate (num_predict on Ollama, --predict in llama.cpp), not the context size (num_ctx on Ollama, --ctx-size in llama.cpp)?
OpenAI API Docs: "The maximum number of tokens that can be generated in the chat completion. The total length of input tokens and generated tokens is limited by the model's context length."

<!-- gh-comment-id:2238282320 --> @chigkim commented on GitHub (Jul 19, 2024): If max_tokens controls the context size, how can we control num_predict with OpenAI API? Isn't max_tokens on OpenAI API supposed to control how many tokens to generate (num_predict on Ollama, --predict in llama.cpp), not the context size (num_ctx on Ollama, --ctx-size in llama.cpp)? [OpenAI API Docs](https://platform.openai.com/docs/api-reference/chat): "The maximum number of tokens that can be generated in the chat completion. The total length of input tokens and generated tokens is limited by the model's context length."
GiteaMirror commented 2026-04-12 11:34:20 -05:00
Author
Owner
Copy Link

@hellopahe commented on GitHub (Jul 23, 2024):

My question is the same to @chigkim 's, It seems like ollama openapi server will truncate input length to default n_ctx=2048, no matter request parameter was brought either by max_tokens or num_ctx.

<!-- gh-comment-id:2244960612 --> @hellopahe commented on GitHub (Jul 23, 2024): My question is the same to @chigkim 's, It seems like ollama openapi server will truncate input length to default `n_ctx=2048`, no matter request parameter was brought either by `max_tokens` or `num_ctx`.
GiteaMirror commented 2026-04-12 11:34:20 -05:00
Author
Owner
Copy Link

@de-code commented on GitHub (Feb 10, 2025):

For anyone coming here looking for the answer. Copying from the aforementioned ollama openai docs:

The OpenAI API does not have a way of setting the context size for a model.

(A PR that that would have added the option of setting the context lengthy via the API was closed)

The current way to change the context length is by creating a model (e.g. via the Modelfile mentioned in the docs)

<!-- gh-comment-id:2647766433 --> @de-code commented on GitHub (Feb 10, 2025): For anyone coming here looking for the answer. Copying from the aforementioned [ollama openai docs](https://github.com/ollama/ollama/blob/main/docs/openai.md): > The OpenAI API does not have a way of setting the context size for a model. (A [PR that that would have added the option of setting the context lengthy via the API](https://github.com/ollama/ollama/pull/8672) was closed) The current way to change the context length is by creating a model (e.g. via the Modelfile mentioned in the docs)
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 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#1633
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?