Skip to content

Conversation

@joetvinson
Copy link

No description provided.

joetvinson and others added 2 commits June 6, 2025 16:18
The `modal_webui.py` script is what I use to deploy the SparkTTS Gradio web UI on Modal.

With this change, I've added the `@modal.concurrent(max_inputs=10)` decorator to the main `run` function. This is in line with Modal's best practices and examples for deploying web server applications (like Gradio, Streamlit, ComfyUI) that are started via `subprocess.Popen`.

Adding this decorator allows the Modal container to accept multiple simultaneous client connections from you (up to 10 in this case), which should improve responsiveness if multiple users are interacting with it. The actual processing of these requests by the SparkTTS model will still depend on the underlying Gradio server and GPU capacity.

I reviewed the script and found it to be otherwise correctly implementing Modal's guidelines for non-ASGI web servers.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant