ComfyUI API v1.0.0

Enterprise GPU Queue & Workflow Gateway
System Online
Queue Depth
-
Pending GPU assignment
Active Processing
-
Inference in progress
Completed Jobs
-
Total finalized runs
GPU Worker Fleet
-
Active worker nodes

Live Execution Queue

Job ID Workflow Status Progress Worker Node Created Actions
Loading queue...

Dispatch New Generation Job

Tip: put file in E:\ComfyUI_windows_portable\ComfyUI\input\ then enter its name, or pick via file picker (name auto-filled).

Registered Workflow Catalog

Loading workflows...

GPU Worker Node Fleet

Loading worker nodes...

API Quickstart & SDK Reference

1. Submit Job via cURL

curl -X POST https://api.comfyui.hardonian.com/api/v1/jobs \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "workflow_id": "wf_flux_portrait",
    "params": {
      "prompt": "futuristic cybernetic cyberpunk portrait, glowing neon, 8k",
      "steps": 28,
      "seed": 1337
    },
    "priority": 8
  }'

2. Python SDK Client

from client.comfyui_api_client import ComfyUIClient

client = ComfyUIClient(api_url="https://api.comfyui.hardonian.com", api_key="YOUR_API_KEY")

# Submit and wait for output artifacts
result = client.generate(
    workflow="flux-portrait-studio",
    params={"prompt": "Studio portrait of a character in rain", "seed": 42},
    priority=8,
    on_progress=lambda p: print(f"Progress: {p['progress']}%")
)
print("Generated artifacts:", result["outputs"])

3. Run Worker Daemon on GPU Node

# On your EPYC GPU Server
python -m bridge.comfy_worker_daemon \
  --api-url https://api.comfyui.hardonian.com \
  --comfy-url http://127.0.0.1:8188 \
  --worker-key YOUR_WORKER_KEY