forked from ggml-org/llama.cpp
-
Notifications
You must be signed in to change notification settings - Fork 618
Closed
Description
Describe the Issue
I'm using kobold to generate images, and I'm trying to build a node based UI, something like ComfyUI.
So my plan is to chain /sdapi/v1/img2img calls with nodes and queues, but between those calls also call Kobold admin API, to switch configs/models.
By posting to /api/admin/reload_config for example {"filename": "model.gguf", "overrideconfig": "config.kcpps}
The issue is that the admin api returns response immediatley, that hey request was successful, but in reality the model switch is still in progress.
Could this be changed, so reload_config would return promise that resolves, once Kobold is ready to be prompted again.
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels