Some Ollama models have small max token limit (~3500 tokens)

[hopeseekr]

What happened?

I'm experiencing it on qwen3-coder and gpt-oss... very small limit. can't fit a 450-line JS file...

The latest message token count is too long, exceeding the token limit, or your token limit parameters are misconfigured, adversely affecting the context window. More info: 3873 / 3686.

3686 tokens is very small.

Version Information

0.7.9 via AUR on Arch Linux.

Steps to Reproduce

1. Install ollama.
2. Download one of the bugged models (gpt-oss:22)
3. Attempt to post a 500 line file of > 3500 tokens.

What browsers are you seeing the problem on?

Chrome

Relevant log output

No error logs in syslog.

Browser error:

The latest message token count is too long, exceeding the token limit, or your token limit parameters are misconfigured, adversely affecting the context window. More info: 3849 / 3686. Please shorten your message, adjust the max context size from the conversation parameters, or fork the conversation to continue.

Screenshots

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

The small context limit is expected as we don't maintain context limits for every model. You can adjust this manually:

1. Use the Parameters panel in the UI to increase max context tokens
2. Or, configure the limit when setting up an agent

For customization of model limits via config, see the existing issue: #1633

You're welcome to submit a PR if you'd like to add specific model limits to the codebase.