Cloud Render API

The Cloud Render API at render.varg.ai lets you submit varg TSX code and get back a rendered video (or image frames). No Bun, no ffmpeg, no local setup — just an API key and curl. This is separate from the Gateway API (api.varg.ai) which generates individual assets (images, videos, speech). The Cloud Render API composes full videos from TSX — calling the Gateway internally for AI generation, then stitching everything together.

Base URL

https://render.varg.ai

Authentication

All requests require a varg API key in the Authorization header:

Authorization: Bearer varg_xxx

Get your API key at app.varg.ai. See Authentication for details.

Quick start

Submit TSX code, poll for the result:

# 1. Submit a render job
JOB=$(curl -s -X POST https://render.varg.ai/api/render \
  -H "Authorization: Bearer $VARG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "code": "import { Render, Clip, Image, Video } from \"vargai/react\";\nimport { fal } from \"vargai/ai\";\nconst img = Image({ prompt: \"cute orange cat, Pixar style\", aspectRatio: \"9:16\" });\nexport default <Render width={1080} height={1920}><Clip duration={3}><Video prompt={{ text: \"cat waves hello\", images: [img] }} model={fal.videoModel(\"kling-v3\")} /></Clip></Render>;"
  }')

JOB_ID=$(echo $JOB | jq -r '.job_id')
echo "Job: $JOB_ID"

# 2. Poll until complete
while true; do
  RESULT=$(curl -s "https://render.varg.ai/api/render/jobs/$JOB_ID" \
    -H "Authorization: Bearer $VARG_API_KEY")
  STATUS=$(echo $RESULT | jq -r '.status')
  echo "Status: $STATUS"
  if [ "$STATUS" = "completed" ] || [ "$STATUS" = "failed" ]; then
    echo $RESULT | jq .
    break
  fi
  sleep 5
done

Cloud render uses fal.*Model() syntax in submitted code — provider globals are auto-injected server-side. You do not need to call createVarg().

Endpoints

POST /api/render

Submit TSX code for rendering. Returns 202 Accepted immediately with a job ID. Request body:

{
  "code": "import { Render, Clip, Image } from \"vargai/react\";\nexport default <Render width={1080} height={1920}><Clip duration={3}><Image prompt=\"sunset\" /></Clip></Render>;",
  "output": "video",
  "mode": "strict",
  "verbose": false,
  "providerKeys": {
    "fal": "fal_xxx",
    "elevenlabs": "el_xxx"
  }
}

Field	Type	Required	Description
`code`	string	Yes	TSX source code. Must export a default `<Render>` element.
`output`	`"video"` or `"frames"`	No	Output format. Default: `"video"`. Use `"frames"` for image-only renders (skips video stitching). Auto-detected if all elements are images.
`mode`	`"strict"` or `"preview"`	No	Render mode. `"strict"` (default) fails on errors. `"preview"` uses placeholders for failed generations.
`verbose`	boolean	No	Enable detailed logging. Default: `false`.
`providerKeys`	object	No	BYOK provider keys. See BYOK.

Response (202):

{
  "job_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "status": "rendering",
  "estimated_duration_ms": 45000,
  "queue": {
    "active": 3,
    "waiting": 1
  }
}

Rate limit headers are included on every response:

X-RateLimit-Limit: 10
X-RateLimit-Remaining: 8
X-RateLimit-Reset: 2026-03-25T12:01:00.000Z

GET /api/render/jobs/:id

Poll a render job’s status. Response (completed):

{
  "job_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "status": "completed",
  "output_url": "https://s3.varg.ai/media/abc123.mp4",
  "thumbnail_url": "https://s3.varg.ai/media/abc123_thumb.jpg",
  "files": [
    {
      "url": "https://s3.varg.ai/media/img_001.png",
      "mediaType": "image/png",
      "metadata": { "type": "image", "model": "fal:nano-banana-pro", "prompt": "cute orange cat" }
    },
    {
      "url": "https://s3.varg.ai/media/vid_001.mp4",
      "mediaType": "video/mp4",
      "metadata": { "type": "video", "model": "fal:kling-v3", "prompt": "cat waves hello" }
    }
  ],
  "duration_ms": 42150,
  "estimated_duration_ms": 45000,
  "created_at": "2026-03-25T12:00:00.000Z",
  "started_at": "2026-03-25T12:00:01.000Z",
  "completed_at": "2026-03-25T12:00:43.000Z"
}

Response (failed):

{
  "job_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "status": "failed",
  "error": "Image generation returned no data for clip 0",
  "error_category": "provider_error",
  "created_at": "2026-03-25T12:00:00.000Z",
  "completed_at": "2026-03-25T12:00:15.000Z"
}

Job statuses:

Status	Description
`rendering`	Job is queued or actively rendering
`completed`	Video/frames are ready at `output_url`
`failed`	Render failed — check `error` field

GET /api/render/jobs/:id/stream

Stream render progress via Server-Sent Events (SSE). Preferred over polling for real-time updates.

curl -N "https://render.varg.ai/api/render/jobs/$JOB_ID/stream" \
  -H "Authorization: Bearer $VARG_API_KEY"

Event format:

event: status
data: {"job_id":"a1b2...","status":"rendering"}

:keepalive

event: status
data: {"job_id":"a1b2...","status":"completed","output_url":"https://s3.varg.ai/media/abc.mp4","files":[...]}

Event	Description
`status`	Job status update. Data is the same JSON as the poll endpoint.
`timeout`	Stream timed out (15 minutes). Reconnect or switch to polling.
`:keepalive`	Heartbeat comment every 15 seconds. Not a real event — keeps the connection alive.

The stream closes automatically when the job reaches a terminal state (completed or failed). If the job is already terminal when you connect, you’ll receive a single status event and the stream closes immediately.

GET /api/jobs

List recent render jobs for the authenticated user. Query parameters:

Param	Type	Default	Description
`limit`	number	50	Max jobs to return (max 200)

Response:

{
  "jobs": [
    {
      "id": "a1b2c3d4-...",
      "status": "completed",
      "outputUrl": "https://s3.varg.ai/media/abc.mp4",
      "durationMs": 42150,
      "createdAt": "2026-03-25T12:00:00.000Z",
      "completedAt": "2026-03-25T12:00:43.000Z"
    }
  ],
  "count": 1
}

Output formats

Video (default)

The standard output. Your TSX is rendered into a single .mp4 file with all clips, transitions, audio, and captions composited together.

{ "output": "video" }

The final video URL is in output_url. Individual generated assets (images, videos, audio) are listed in the files array.

Frames

For image-only renders (no video/audio elements), the service can skip video stitching and return individual image frames. This is faster and useful for generating image sets.

{ "output": "frames" }

If your TSX contains only <Image> elements (no <Video>, <Speech>, <Music>, <Captions>), frames mode is auto-detected even without specifying it. Frame results appear in the files array with clipIndex for ordering.

BYOK (Bring Your Own Keys)

Pass your own provider API keys to bypass varg billing. The render service forwards them to the Gateway for AI generation:

{
  "code": "...",
  "providerKeys": {
    "fal": "fal_xxx",
    "elevenlabs": "el_xxx",
    "higgsfield": "hf_xxx",
    "replicate": "r8_xxx",
    "openai": "sk_xxx",
    "google": "goog_xxx"
  }
}

See the BYOK guide for details.

TSX code requirements

The submitted code must:

Import from vargai/react — Render, Clip, Image, Video, Speech, Music, Captions, Title, etc.
Import providers from vargai/ai — fal, elevenlabs, higgsfield, replicate (these are auto-injected as globals)
Export a default <Render> element — this is the root of your video composition

import { Render, Clip, Image, Video, Speech, Captions, Music } from "vargai/react";
import { fal, elevenlabs } from "vargai/ai";

const character = Image({
  prompt: "friendly tech reviewer, casual style",
  model: fal.imageModel("nano-banana-pro"),
  aspectRatio: "9:16",
});

const voiceover = Speech({
  model: elevenlabs.speechModel("eleven_v3"),
  voice: "josh",
  children: "Hey everyone! Let me show you something cool.",
});

export default (
  <Render width={1080} height={1920}>
    <Music prompt="upbeat tech vlog" model={elevenlabs.musicModel()} volume={0.1} />
    <Clip duration={5}>
      <Video
        prompt={{ text: "person talking naturally", images: [character] }}
        model={fal.videoModel("kling-v3")}
      />
    </Clip>
    <Captions src={voiceover} style="tiktok" color="#ffffff" />
  </Render>
);

Cloud render uses fal.*Model(), elevenlabs.*Model(), etc. — not varg.*Model(). Provider globals are injected server-side. The createVarg() pattern is for local rendering only.

Rate limits

Limit	Value
Concurrent renders per user	5
Requests per minute	10
Stream timeout	15 minutes

When rate limited, the response includes a Retry-After: 30 header.

Error codes

Status	Error	Description
400	`Invalid JSON body`	Request body is not valid JSON
400	`Invalid render code`	TSX code has syntax errors or doesn’t export a valid element
401	`Unauthorized`	Missing or invalid API key
429	Rate limit	Too many concurrent renders or requests per minute
503	`Failed to enqueue render job`	Queue is unavailable

Gateway API vs Cloud Render API

	Gateway API	Cloud Render API
URL	`api.varg.ai/v1`	`render.varg.ai`
Purpose	Generate individual assets (image, video, speech, music)	Compose full videos from TSX code
Input	JSON with prompt/model	TSX source code
Output	Single asset URL	Rendered video + all intermediate assets
Auth	`Authorization: Bearer varg_xxx`	Same
Use when	You need one image or video clip	You need a complete, composed video with transitions, audio, captions

Overview

Generation

Jobs

Files

Reference

Account

Cloud Render API

Base URL

Authentication

Quick start

Endpoints

POST /api/render

GET /api/render/jobs/:id

GET /api/render/jobs/:id/stream

GET /api/jobs

Output formats

Video (default)

Frames

BYOK (Bring Your Own Keys)

TSX code requirements

Rate limits

Error codes

Gateway API vs Cloud Render API

Overview

Generation

Jobs

Files

Reference

Account

​Base URL

​Authentication

​Quick start

​Endpoints

​POST /api/render

​GET /api/render/jobs/:id

​GET /api/render/jobs/:id/stream

​GET /api/jobs

​Output formats

​Video (default)

​Frames

​BYOK (Bring Your Own Keys)

​TSX code requirements

​Rate limits

​Error codes

​Gateway API vs Cloud Render API

Base URL

Authentication

Quick start

Endpoints

POST /api/render

GET /api/render/jobs/:id

GET /api/render/jobs/:id/stream

GET /api/jobs

Output formats

Video (default)

Frames

BYOK (Bring Your Own Keys)

TSX code requirements

Rate limits

Error codes

Gateway API vs Cloud Render API