Skip to main content
GET
https://api.comfycontrol.app
/
v1
/
runner
/
{id}
/
stats
Get Runner Stats
curl --request GET \
  --url https://api.comfycontrol.app/v1/runner/{id}/stats \
  --header 'Authorization: Bearer <token>'
{
  "401": {},
  "404": {},
  "500": {},
  "system": {},
  "system.os": "<string>",
  "system.ram_free": 123,
  "system.ram_total": 123,
  "system.comfyui_version": "<string>",
  "system.python_version": "<string>",
  "devices": [
    {}
  ],
  "devices[].name": "<string>",
  "devices[].type": "<string>",
  "devices[].vram_total": 123,
  "devices[].vram_free": 123,
  "dev_mode": true,
  "extensions": [
    "<string>"
  ]
}

​
Overview

Fetch real-time system statistics from a ComfyUI runner, including hardware information, resource usage, installed extensions, and configuration settings.
This endpoint makes a live request to your ComfyUI instance, so response time depends on your runner’s network latency.

​
Path Parameters

​
id
string
required
The unique identifier (UUID) of the runner to query.

​
Response

​
system
object
required
System information.
​
system.os
string
Operating system name.
​
system.ram_free
integer
Free RAM in bytes.
​
system.ram_total
integer
Total RAM in bytes.
​
system.comfyui_version
string
ComfyUI version number.
​
system.python_version
string
Python version running ComfyUI.
​
devices
array
required
Array of compute devices (GPUs/CPUs) available to ComfyUI.
​
devices[].name
string
Device name/model.
​
devices[].type
string
Device type (e.g., “cuda”, “cpu”, “mps”).
​
devices[].vram_total
integer
Total VRAM in bytes (for GPUs).
​
devices[].vram_free
integer
Free VRAM in bytes (for GPUs).
​
dev_mode
boolean
required
Whether ComfyUI is running in development mode.
​
extensions
string[]
required
Array of installed ComfyUI extensions/custom nodes.

​
Example Response

{
  "system": {
    "os": "posix",
    "ram_free": 239499096064,
    "ram_total": 253344550912,
    "comfyui_version": "0.3.77",
    "python_version": "3.12.12 (main, Oct 15 2025, 09:20:08) [GCC]"
  },
  "devices": [
    {
      "name": "cuda:0 NVIDIA H100 80GB HBM3 : cudaMallocAsync",
      "type": "cuda",
      "vram_total": 84942848000,
      "vram_free": 84316848128
    }
  ],
  "dev_mode": true,
  "extensions": [
    "/extensions/core/groupNodeManage.js",
    "/extensions/core/groupNode.js",
    "/extensions/ComfyUI-Manager/comfyui-manager.js",
    "/extensions/rgthree-comfy/rgthree.js"
  ]
}

​
Error Responses

​
401
error
Unauthorized - invalid or missing authentication token.
​
404
error
Not Found - runner with the specified ID does not exist or does not belong to your account.
​
500
error
Internal Server Error - unable to connect to the runner or retrieve stats.Common causes:
  • Runner endpoint is unreachable
  • Runner is offline or not responding
  • Network connectivity issues
  • Invalid authentication headers configured for the runner

​
Notes

  • This endpoint makes a live request to your ComfyUI instance
  • Response time varies based on network latency
  • VRAM and RAM values are in bytes (divide by 1,073,741,824 for GB)
  • Use this to monitor runner health and capacity before scheduling workflows
  • System information includes ComfyUI version and Python version details