Use Ray Flash 2 with Comfy Router

Call luma/ray-flash-2 through Comfy Router: endpoint, request shape and the response Router returns.

API Reference for luma/ray-flash-2, served by Comfy Router from Luma.

Quick start

Create a key in your Comfy workspace and export it as COMFY_API_KEY. The Python and TypeScript snippets use the Comfy SDKs (pip install comfy-sdk, npm install @comfyorg/sdk); the cURL snippet is the same call over raw HTTP.

Model ID: luma/ray-flash-2

Endpoint: POST https://api.comfy.org/v2/models/luma/ray-flash-2

Python

from comfy_sdk import Comfy

# Reads COMFY_API_KEY from the environment.
# The SDK automatically creates an idempotency key and reuses it for automatic retries.
with Comfy() as client:
    result = client.models.run(
        "luma/ray-flash-2",
        {
            "aspect_ratio": "16:9",
            "duration": "5s",
            "prompt": "a single red maple leaf resting on a plain white background",
            "resolution": "540p",
        },
    )

print(result)

Schema

Input

aspect_ratio · string · default: "\"16:9\"" · required

The aspect ratio of the generation

Possible values: 1:1, 16:9, 9:16, 4:3, 3:4, 21:9, 9:21

callback_url · string (uri)

The callback URL of the generation, a POST request with Generation object will be sent to the callback URL when the generation is dreaming, completed, or failed

Format: uri

duration · `5s`, `9s` | string · required

generation_type · string · default: "\"video\""

Possible values: video

keyframes · object

The keyframes of the generation

keyframes.frame0 · object

A keyframe can be either a Generation reference, an Image, or a Video

keyframes.frame1 · object

A keyframe can be either a Generation reference, an Image, or a Video

loop · boolean

Whether to loop the video

model · string

The video model used for the generation. On the Comfy Router route POST /v2/models/luma/{model} this field is supplied from the path and MUST NOT be sent; on the v1 POST /proxy/luma/generations route it is required and constrained to the LumaVideoModel enum (ray-2, ray-flash-2).

prompt · string · required

The prompt of the generation

resolution · `540p`, `720p`, `1080p`, `4k` | string · required

Generated from the schema Router serves at GET /v2/models/luma/ray-flash-2/openapi.json, the same document it validates a call against before the request reaches the provider.

Output

assets · object

The assets of the generation

assets.image · string (uri)

The URL of the image

Format: uri

assets.progress_video · string (uri)

The URL of the progress video

Format: uri

assets.video · string (uri)

The URL of the video

Format: uri

created_at · string (date-time)

The date and time when the generation was created

Format: date-time

failure_reason · string

The reason for the state of the generation

generation_type · string

Possible values: video, image

id · string (uuid)

The ID of the generation

Format: uuid

model · string

The model used for the generation

request · object

The request of the generation

state · string

The state of the generation

Possible values: queued, dreaming, completed, failed

Examples

Input

{
  "aspect_ratio": "16:9",
  "duration": "5s",
  "prompt": "a single red maple leaf resting on a plain white background",
  "resolution": "540p"
}

Output

{
  "assets": {
    "video": "https://example.invalid/luma/ray-flash-2/generated.mp4"
  },
  "created_at": "2027-01-01T00:00:00Z",
  "generation_type": "video",
  "id": "8c41d0b6-2e7a-4f35-b1d8-9a0e6f3c2b71",
  "model": "ray-flash-2",
  "state": "completed"
}

Before you ship

The SDKs create an Idempotency-Key and reuse it for automatic retries. For manual retries, reuse the original key. Router can hold the connection for up to 10 minutes.

When a request fails, Router sends an X-Comfy-Error-Type response header explaining why. A 422 means Router rejected the input before calling the provider. Download generated assets promptly because result URLs can expire.

Headers

Authentication, idempotency, request IDs, error buckets, retry pacing, spend limits.

Using the Router API

Model discovery, validation errors, retries, and billing.

Limitations

What Router does not do today, and what to use instead.