GET /v1/models

Lists the hosted models, what each costs, and whether it's available right now. Free; needs an API key.

curl https://openeye.sh/v1/models \
  -H "Authorization: Bearer oe_your_key_here"

[
  { "id": "yolov8", "name": "YOLOv8 Nano", "task": "detection", "credits_per_call": 1,
    "description": "Real-time object detection with 80 COCO classes.", "available": true },
  { "id": "qwen2.5-vl", "name": "Qwen2.5-VL", "task": "description", "credits_per_call": 3,
    "description": "Scene description and visual Q&A.", "available": false },
  { "id": "depth-anything-v2", "name": "Depth Anything V2", "task": "depth", "credits_per_call": 2,
    "description": "Monocular depth estimation. Self-hosted only for now (openeye serve).", "available": false }
]

Model	Endpoint	Status
yolov8	POST /v1/detect	Available
qwen2.5-vl	POST /v1/describe	Becomes available when the VLM is enabled; returns 503 until then
depth-anything-v2	POST /v1/depth	Not hosted; returns 501

From the CLI: openeye api models. For models on your own machine, run openeye list.

← POST /v1/describe GET /v1/usage →