bug: Issues with cli on Linux and Mac

[avb-is-me]

Was able to globally install the CLI, and complete a pull, but found on both linux(Github codespaces 16GB) and Mac(Apple m2 8G), ran into issues with cortex models run llama3. When I ran the command it seemed like nothing happened. Additionally serve would run successfully but when visited lead to a 404.

Looks Successful

When visited:

To Reproduce

Steps to reproduce the behavior, follow these instructions:

1. Install Cortex using NPM

npm i -g @janhq/cortex

2. Download a GGUF model

cortex models pull llama3

3. Run the model to start chatting

cortex models run llama3

4. (Optional) Run Cortex in OpenAI-compatible server mode

cortex serve

Expected behavior
Success message or interaction on run {model}, and swagger ui on cortex serve

Desktop (please complete the following information):

• OS: Linux, Mac

[l2D]

Edit:

I tried to run cortex run <model_id> and it seem working correctly.

cortex run llama3
Inorder to exit, type 'exit()'.
>> tell me a joke
Here's one:

Why couldn't the bicycle stand up by itself?

(wait for it...)

Because it was two-tired!

Hope that made you laugh!

>>

My device:

• Chip: Apple M3 Pro
• Memory: 18 GB
• OS: Sonoma 14.5 (23F79)

---

You should run the model with:

cortex models start <model_id>

# Output:
# cortex models start llama3
# { message: 'Model loaded successfully', modelId: 'llama3' }

and Swagger UI available at http://${host}:${port}/api e.g. http://localhost:1337/api

Screenshot

[avb-is-me]

@louis-jan did you try on a github codespaces using linux? Also I think try a fresh install, aka for users that haven't installed cortex yet or pulled a model