Provider Templates

Provider templates let you integrate with popular AI APIs using a single provider field instead of writing custom polling configuration. AsyncQueue knows how to submit, poll, and collect results from each supported provider.

Available Providers

ID	Provider	Description	Poll Interval	Max Duration
fal	fal.ai	Image and video generation	3s	10 min
replicate	Replicate	Open-source AI models	5s	10 min
runwayml	RunwayML	Video generation (Gen-3, Gen-4)	5s	10 min
luma	Luma AI	Dream Machine video generation	5s	10 min
elevenlabs-dubbing	ElevenLabs	Video/audio dubbing and translation	10s	30 min
stability-video	Stability AI	Stable Video Diffusion	10s	10 min

Listing Providers via API

curl https://api.asyncqueue.io/v1/providers

Returns all available templates with their full pollConfig objects.

---

fal.ai

Generate images and video with fal.ai models (Flux, Kling, Minimax, and more).

fal.ai returns status_url and response_url in the submit response. AsyncQueue uses these directly for polling and result fetching.

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "X-API-Key: async_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://queue.fal.run/fal-ai/flux/dev",
    "method": "POST",
    "headers": {
      "Authorization": "Key fal_your_key"
    },
    "body": "{\"prompt\": \"A cat riding a skateboard\", \"image_size\": \"landscape_16_9\"}",
    "provider": "fal",
    "onCompleteUrl": "https://your-app.com/api/image-ready"
  }'

Status values: IN_QUEUE, IN_PROGRESS, COMPLETED

Result: Full model output fetched from the separate response_url.

---

Replicate

Run any open-source model on Replicate (Stable Diffusion, Whisper, LLaMA, video models, and thousands more).

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "X-API-Key: async_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://api.replicate.com/v1/models/stability-ai/stable-diffusion-3/predictions",
    "method": "POST",
    "headers": {
      "Authorization": "Bearer r8_your_replicate_token"
    },
    "body": "{\"input\": {\"prompt\": \"An astronaut riding a horse\"}}",
    "provider": "replicate",
    "onCompleteUrl": "https://your-app.com/api/prediction-ready"
  }'

Status values: starting, processing, succeeded, failed, canceled

Result: The output field from the prediction response (typically an array of URLs for image models).

---

RunwayML

Generate video with RunwayML Gen-3, Gen-4, and Gen-4.5 models. RunwayML does not support webhooks, so polling is the only option.

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "X-API-Key: async_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://api.dev.runwayml.com/v1/image_to_video",
    "method": "POST",
    "headers": {
      "Authorization": "Bearer runway_your_key",
      "X-Runway-Version": "2024-11-06"
    },
    "body": "{\"model\": \"gen4\", \"promptImage\": \"https://example.com/image.jpg\", \"promptText\": \"A timelapse on a sunny day\", \"duration\": 5}",
    "provider": "runwayml",
    "onCompleteUrl": "https://your-app.com/api/video-ready"
  }'

Status values: PENDING, THROTTLED, RUNNING, SUCCEEDED, FAILED

Result: The output array (typically contains a video URL). Video URLs expire within 24-48 hours - download them promptly.

---

Luma AI

Generate video with Luma AI’s Dream Machine (Ray-2 model).

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "X-API-Key: async_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://api.lumalabs.ai/dream-machine/v1/generations",
    "method": "POST",
    "headers": {
      "Authorization": "Bearer luma_your_key"
    },
    "body": "{\"prompt\": \"A drone flyover of a mountain lake at sunrise\", \"model\": \"ray-2\"}",
    "provider": "luma",
    "onCompleteUrl": "https://your-app.com/api/video-ready"
  }'

Status values: dreaming, completed, failed

Result: The assets.video field (a URL to the generated video).

---

ElevenLabs Dubbing

Dub and translate video or audio content. Dubbing jobs can take several minutes depending on content length.

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "X-API-Key: async_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://api.elevenlabs.io/v1/dubbing",
    "method": "POST",
    "headers": {
      "xi-api-key": "el_your_key",
      "Content-Type": "application/json"
    },
    "body": "{\"source_url\": \"https://example.com/video.mp4\", \"target_lang\": \"es\"}",
    "provider": "elevenlabs-dubbing",
    "onCompleteUrl": "https://your-app.com/api/dubbing-ready"
  }'

Status values: dubbing, dubbed, failed

Result: The dubbing_id - use it to fetch the dubbed audio via GET /v1/dubbing/{dubbing_id}/audio/{language_code} on ElevenLabs’ API.

---

Stability AI Video

Generate video from images using Stable Video Diffusion.

Note: Stability AI uses HTTP status codes instead of JSON status fields. A 202 response means still processing, 200 means complete.

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "X-API-Key: async_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://api.stability.ai/v2beta/image-to-video",
    "method": "POST",
    "headers": {
      "Authorization": "Bearer sk-your-stability-key"
    },
    "body": "{\"image\": \"base64_encoded_image_data\", \"seed\": 42}",
    "provider": "stability-video",
    "onCompleteUrl": "https://your-app.com/api/video-ready"
  }'

Status values: HTTP 200 (complete), HTTP 202 (in progress)

Result: The binary video content URL.

---

Custom Providers

For AI APIs not in the template list, provide a custom pollConfig object instead of a provider name. See AI Provider Polling for the full pollConfig reference and template variable syntax.