Skip to content

README.md

Latest commit

 

History

History
47 lines (33 loc) · 1000 Bytes

README.md

File metadata and controls

47 lines (33 loc) · 1000 Bytes

fastapi-comfyui-orchestrator

The cognitive service container submits ComfyUI workflows to its host running server. Reproducible with Omnigen2 Text-to-Image.

screenshot

Requirements

git (test v: 2.34.1)
docker (test v: 28.4.0)
docker-compose (test v: 2.39.1)

Development and Setup

./run_all.sh

# Service health
curl -sS http://localhost:8001/api/cognitive/health

# Queue a workflow
curl -sS -X POST http://localhost:8001/api/cognitive/comfyui \
	-H 'Content-Type: application/json' \
	-d '{"prompt":"A cat with a crown lounging on a velvet throne","workflow_name":"image_omnigen2_t2i"}'

# Poll workflow status
curl -sS http://localhost:8001/api/cognitive/comfyui/<prompt_id>

API

# Cognitive service (port 8001)
GET  /api/cognitive/health
POST /api/cognitive/comfyui
GET  /api/cognitive/comfyui/{prompt_id}

# Root UI
GET /

# ComfyUI host API used by cognitive (port 8188)
POST /prompt
GET  /history/{prompt_id}