Use LTX 2.5 Fast with Comfy Router

Call ltx/ltx-2-5-fast through Comfy Router: endpoint, request shape and the response Router returns.

API Reference for ltx/ltx-2-5-fast, served by Comfy Router from LTX.

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: ltx/ltx-2-5-fast

Endpoint: POST https://api.comfy.org/v2/models/ltx/ltx-2-5-fast

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(
        "ltx/ltx-2-5-fast",
        {
            "duration": 2,
            "fps": 24,
            "generate_audio": False,
            "prompt": "A single red maple leaf resting on a plain white background.",
            "resolution": "1280x720",
        },
    )

print(result)

Schema

Input

duration · integer · required

Video duration in seconds (maximum depends on resolution and frame rate)

Possible values: 2, 3, 4, 5, 6, 8, 10, 12, 14, 16, 18, 20

fps · integer · default: "25"

Frame rate in frames per second

Possible values: 24, 25, 48, 50

generate_audio · boolean · default: "true"

Generate audio for the video

model · string

Model to use for generation. On the Comfy Router route POST /v2/models/ltx/{model} this field is supplied from the path and may be omitted. The spellings LTX serves on this operation — the set Comfy Router addresses as ltx/<model> — are ltx-2-5-fast and ltx-2-5-pro; they are written out here rather than constrained to an enum for the reason the comment above gives. Which resolutions each one admits is the x-comfy-model-resolutions matrix on the resolution property below.

prompt · string · required

Text prompt describing the desired video content

resolution · string · required

Output video resolution. The enum is the union over all models; the supported set is per model. Supported pairs: ltx-2-5-fast: 1280x720, 720x1280, 1920x1080, 1080x1920, 2560x1440, 1440x2560, 3840x2160, 2160x3840; ltx-2-5-pro: 1280x720, 720x1280, 1920x1080, 1080x1920. Other (model, resolution) pairs are not supported; the v2 routes reject them with 400. The same matrix is published machine-readably in this property’s x-comfy-model-resolutions extension.

Possible values: 1280x720, 720x1280, 1920x1080, 1080x1920, 2560x1440, 1440x2560, 3840x2160, 2160x3840

Generated from the schema Router serves at GET /v2/models/ltx/ltx-2-5-fast/openapi.json, the same document it validates a call against before the request reaches the provider.

Output

completed_at · string

Job completion timestamp (ISO 8601)

created_at · string

Job creation timestamp (ISO 8601)

error · object

Present when status is failed

error.message · string

error.type · string

id · string

Unique job identifier

result · object

Present when status is completed; output URLs expire 24 hours after completion

result.video_url · string

URL of the generated video

status · string

Job status (pending, processing, completed, failed)

Examples

Input

{
  "duration": 2,
  "fps": 24,
  "generate_audio": false,
  "prompt": "A single red maple leaf resting on a plain white background.",
  "resolution": "1280x720"
}

Output

{
  "completed_at": "2026-01-01T00:02:10Z",
  "created_at": "2026-01-01T00:00:00Z",
  "id": "3f7a1b28-5c0d-4e91-8a6f-1b2c3d4e5f60",
  "result": {
    "video_url": "https://example.invalid/ltx/generated.mp4"
  },
  "status": "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.