lmstudio-ai / lmstudio-ai/lmstudio-python

Does .chat() support setting context length?

Open
#173 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
875
Forks
166
PR merge metrics
No merged PRs in 30d

Description

Hello I'm trying to set context length when running this code. Placing the configuration value in different places hasn't worked.

I am able to load the model with manually set context length in the GUI, then the terminal program runs fine using model = lms.llm()

https://lmstudio.ai/docs/python/llm-prediction/chat-completion#example-multi-turn-chat

If I load the config value with the model it doesn't apply the setting when opening .chat()
model = lms.llm("OpenAI-20B-NEO-CODEPlus-Uncensored-IQ4_NL.gguf", config={ "contextLength": 8192 })

If I set it in the chat I get an error. Example code:
chat = lms.Chat(systemprompt, config={ "contextLength": 8192 })
error:
line 59, in <module> chat = lms.Chat(systemprompt, config={ "contextLength": 8192 }) TypeError: Chat.__init__() got an unexpected keyword argument 'config'

Help with this issue would be appreciated.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the linked multi-turn chat example and inspect how lms.llm() and lms.Chat() accept configuration. Reproduce the TypeError with the shown contextLength value and determine the supported configuration path; done means the documented chat flow can set a context length of 8192 without the constructor error.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
api
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.