varg API

Every model for video, image and voice.
One key, one request shape.

Seedance, Kling, Veo, Wan, FLUX, Nano Banana, GPT Image, ElevenLabs and the rest behind one REST API — plus a renderer that stitches them into finished videos. Pay per generation from one credit balance.

Get an API key Explore models Docs

Base URLhttps://api.varg.ai/v2

models: 157
video: 75
image: 54
voice & music: 22
providers: 14

Featured

The models people ship with most. Every clip is real output.

Seedance 2.5

ByteDance Seedance 2.5 — the flagship Seedance model. High-quality video from text, with optional image (≤9), video (≤3), and audio (≤3) references. 480p/720p, 4-30s.

from 15.4 credits/sTry it

FLUX 3

Black Forest Labs FLUX 3. Latest flagship video model — text-to-video, image-to-video, and video-to-video with native audio. 720p/1080p, 5-10s.

from 17.9 credits/sTry it

Minimax H3

MiniMax H3. Frontier 2K video model with native stereo audio. Text, image, or multimodal reference inputs.

from 5.3 credits/sTry it

Seedance 2

ByteDance Seedance 2. High-quality video from text, images, or video editing.

from 7.4 credits/sTry it

FLUX 3 Draft

Black Forest Labs FLUX 3 Draft. Fast ~1/3 cost preview mode for FLUX 3 — text-to-video, image-to-video, video-to-video. HD only, 5-10s.

6.3 credits/sTry it

Nano Banana 2

Google Gemini 2 Flash Image. Fast, cost-effective image generation.

~2 creditsTry it

Kling v3

Latest Kling model. Pro tier. Text-to-video and image-to-video.

from 11.8 credits/sTry it

Happy Horse

Alibaba video generation. Expressive character motion.

from 14.7 credits/sTry it

All models

Open any model to try it in the Playground, read its inputs and copy the call. Prices are in credits, from the cheapest route. llms.txt

Seedance 2.5

from 15.4 credits/s

ByteDance Seedance 2.5 — the flagship Seedance model. High-quality video from text, with optional image (≤9), video (≤3), and audio (≤3) references. 480p/720p, 4-30s.

Text to video · Image to video · Video to video

FLUX 3

from 17.9 credits/s

Black Forest Labs FLUX 3. Latest flagship video model — text-to-video, image-to-video, and video-to-video with native audio. 720p/1080p, 5-10s.

Text to video · Image to video · Video to video

Minimax H3

from 5.3 credits/s

MiniMax H3. Frontier 2K video model with native stereo audio. Text, image, or multimodal reference inputs.

Text to video · Image to video

Seedance 2

from 7.4 credits/s

ByteDance Seedance 2. High-quality video from text, images, or video editing.

Text to video · Image to video · Video to video

FLUX 3 Draft

Black Forest Labs FLUX 3 Draft. Fast ~1/3 cost preview mode for FLUX 3 — text-to-video, image-to-video, video-to-video. HD only, 5-10s.

Text to video · Image to video · Video to video

Nano Banana 2

Google Gemini 2 Flash Image. Fast, cost-effective image generation.

Text to image · Image to image

Kling v3

from 11.8 credits/s

Latest Kling model. Pro tier. Text-to-video and image-to-video.

Image to video · Reference to video · Text to video

Happy Horse

from 14.7 credits/s

Alibaba video generation. Expressive character motion.

Text to video · Image to video · Video to video

Ffmpeg

FFmpeg

Flux 3 Image Klein

Black Forest Labs FLUX 3 Image Klein. Compact open-weights FLUX 3 Image (5.6B): native 4K, bounding-box layout, edits with up to 10 references.

Text to image · Image to image

Flux Dev

Flux Dev. Good quality at lower cost.

Text to image

Flux Pro

Black Forest Labs Flux Pro 1.1. High-quality image generation.

Text to image

Flux Schnell

0.315 credits/MP

Flux Schnell. Fastest and cheapest Flux variant.

Text to image

Flux Schnell (PiAPI)

0.315 credits/MP

Flux Schnell via PiAPI provider. Alternative routing for Flux Schnell.

Text to image

Grok Imagine Image

3.2 credits/image

xAI Grok image generation. Cheapest image model.

Text to image

Grok Imagine Image Edit

3.2 credits/image

xAI Grok image editing. Cheapest image editing model.

Image to image

Grok Imagine Video

xAI Grok video generation. Text-to-video and image-to-video.

Text to video · Image to video

Ideogram 4.5

Posters, logos and designs with accurate text. Attach an image to edit it — up to 4 references, an optional mask, and a high-precision mode that keeps untouched pixels.

Text to image · Image to image

Kling v2

from 11.8 credits/s

Master tier. Stable and well-tested.

Text to video · Image to video

Kling v2.5

from 11.8 credits/s

Turbo variant. Fast generation with good quality.

Text to video · Image to video

Kling v2.6

from 11.8 credits/s

Pro tier with improved motion consistency.

Text to video · Image to video

Kling v3 Pro Reference To Video

from 11.8 credits/s

Reference to video

Kling v3 Standard

from 11.8 credits/s

Kling O3 standard tier. Good balance of quality and cost.

Text to video · Image to video

Kling v3 Standard V2V Reference

Video to video

More models 104

More than a model router

The parts you would otherwise build around the models.

Render finished videos

Describe scenes in TSX — clips, voiceover, music, captions — and POST /v2/render generates every asset and stitches the MP4.

Understand a video

A timeline of what happens on screen, the transcript and on-screen text, as JSON.

video_analyze

FFmpeg in the cloud

Trim, slice, resize and run any FFmpeg command on files by URL. No servers to keep.

Webhooks or polling

Every call returns a job at once. Poll it, or pass options.webhook_url and get called back.

options.webhook_url

Result cache

Send the same request twice and the second answer comes from cache instead of a new generation.

options.cache

For agents: MCP

Connect Claude, Cursor or any MCP client to every model. Sign in with varg, no key to paste.

mcp.varg.ai/mcp

Switch models by changing one string

Every tool takes the same shape: a model, a prompt and its options. The call answers with a job right away; poll it or take a webhook. The output lands on the CDN and stays in your files.

One key for every provider
JSON Schema for every model, no key needed
TypeScript SDK, Python SDK and agent skills

TypeScript SDK Agent skills

POST /v2/tools/video/call

curl -X POST https://api.varg.ai/v2/tools/video/call \
  -H "Authorization: Bearer $VARG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "seedance_2_5",
  "prompt": "A red paper lantern drifts up through a rainy night market, neon reflections",
  "duration": 5,
  "resolution": "720p",
  "aspect_ratio": "16:9"
}'

GET /v2/jobs/:id

curl https://api.varg.ai/v2/jobs/$JOB_ID \
  -H "Authorization: Bearer $VARG_API_KEY"

# Poll until "status" is completed, failed or cancelled.
# The result is in output.outputs[0].url

Make the first request in a minute

Create a key, export it as VARG_API_KEY, and send any request above. Keys, usage and billing live in your API console.

Get an API key Open the Playground