Use Photon Flash 1 with Comfy Router

Call luma/photon-flash-1 through Comfy Router: endpoint, request shape and the response Router returns.

API Reference for luma/photon-flash-1, 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/photon-flash-1

Endpoint: POST https://api.comfy.org/v2/models/luma/photon-flash-1

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/photon-flash-1",
        {
            "aspect_ratio": "1:1",
            "prompt": "a red circle on a plain white background",
        },
    )

print(result)

Schema

Input

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

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 for the generation

Format: uri

character_ref · object

character_ref.identity0 · object

The image identity object

character_ref.identity0.images · string (uri)[]

The URLs of the image identity

generation_type · string · default: "\"image\""

Possible values: image

image_ref · object[]

The image reference object

image_ref[].url · string (uri)

The URL of the image reference

Format: uri

image_ref[].weight · number

The weight of the image reference

model · string

The image 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/image route it is constrained to the LumaImageModel enum (photon-1, photon-flash-1), which the Rewrite enforces by refusing an omitted or unknown model.

modify_image_ref · object

The modify image reference object

modify_image_ref.url · string (uri)

The URL of the image reference

Format: uri

modify_image_ref.weight · number

The weight of the modify image reference

prompt · string

The prompt of the generation

style_ref · object[]

The image reference object

style_ref[].url · string (uri)

The URL of the image reference

Format: uri

style_ref[].weight · number

The weight of the image reference

Generated from the schema Router serves at GET /v2/models/luma/photon-flash-1/openapi.json, the same document it validates a call against before the request reaches the provider.

Output

assets · object

The assets of an image generation. image is the result; video and progress_video belong to the Dream Machine VIDEO operation, which shares this status route, and Luma sends them here as explicit nulls.

assets.image · string (uri)

The URL of the generated image — the leaf a luma/photon-* generation fills. Null until the generation reaches completed.

Format: uri

assets.progress_video · string (uri)

Always null on an image generation; it is the in-flight preview Luma publishes while a VIDEO generation runs.

Format: uri

assets.video · string (uri)

Always null on an image generation; the luma/ray-* video models fill it instead.

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

The type of the generation — always image on this operation

Possible values: image

id · string (uuid)

The ID of the generation

Format: uuid

model · string

The model used for the generation

request · object

The image generation request, echoed back inside the terminal document. Luma serialises every field of its request model, writing an explicit null into each one the caller did not set, so read a field’s presence from its VALUE rather than from the key.

request.aspect_ratio · string · default: "\"16:9\""

The aspect ratio of the generation

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

request.callback_url · string (uri)

The callback URL for the generation

Format: uri

request.character_ref · object

request.character_ref.identity0 · object

An image identity as it comes BACK inside the terminal document. Same shape as LumaImageIdentity, with its members nullable for the reason LumaImageRefEcho gives.

request.character_ref.identity0.images · string (uri)[]

The URLs of the image identity

request.generation_type · string

Always image when set, echoing the operation. Null when the caller did not send one, for the reason this whole echo object exists.

request.image_ref · object[]

An image reference as it comes BACK inside the terminal document. Same shape as LumaImageRef, with each member nullable because Luma echoes an unset one as an explicit null rather than omitting it.

request.image_ref[].url · string (uri)

The URL of the image reference

Format: uri

request.image_ref[].weight · number

The weight of the image reference

request.model · string · default: "\"photon-1\""

The image model used for the generation

Possible values: photon-1, photon-flash-1

request.modify_image_ref · object

A modify-image reference as it comes BACK inside the terminal document. Same shape as LumaModifyImageRef, with its members nullable for the reason LumaImageRefEcho gives.

request.modify_image_ref.url · string (uri)

The URL of the image reference

Format: uri

request.modify_image_ref.weight · number

The weight of the modify image reference

request.prompt · string

The prompt of the generation

request.style_ref · object[]

An image reference as it comes BACK inside the terminal document. Same shape as LumaImageRef, with each member nullable because Luma echoes an unset one as an explicit null rather than omitting it.

request.style_ref[].url · string (uri)

The URL of the image reference

Format: uri

request.style_ref[].weight · number

The weight of the image reference

state · string

The state of the generation

Possible values: queued, dreaming, completed, failed

Examples

Input

{
  "aspect_ratio": "1:1",
  "prompt": "a red circle on a plain white background"
}

Output

{
  "assets": {
    "image": "https://example.invalid/luma/photon-flash-1/generated.png",
    "progress_video": null,
    "video": null
  },
  "created_at": "2027-01-01T00:00:00Z",
  "failure_reason": null,
  "generation_type": "image",
  "id": "a1c73f26-08d5-4e69-b7a2-3c4d5e6f7a80",
  "model": "photon-flash-1",
  "request": {
    "aspect_ratio": "1:1",
    "callback_url": null,
    "character_ref": null,
    "generation_type": "image",
    "image_ref": null,
    "model": "photon-flash-1",
    "modify_image_ref": null,
    "prompt": "a red circle on a plain white background",
    "style_ref": null
  },
  "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.