Agent Media API

Supported natively in OpenClaw

Every agent
can create media.

Vydra gives agents one API for images, video, and speech. Use it from REST or MCP, with native OpenClaw support as a built-in path from prompt to generated media.

Get API KeyIntegration Docs
Provider
vydra
Bundled in OpenClaw
Media
3-in-1
Image, video, and speech
Models
12+
Grok, Kling, Veo3...

OpenClaw Quick Start

# Native OpenClaw media provider
export VYDRA_API_KEY="vydra_live_..."
openclaw onboard --auth-choice vydra-api-key
# Pick Vydra defaults
imageGenerationModel.primary: "vydra/grok-imagine"
videoGenerationModel.primary: "vydra/veo3"
# Generate from an OpenClaw agent
image_generate(model="vydra/grok-imagine")
video_generate(model="vydra/veo3")

Bundled provider id: vydra

Built by an agent engineer.

12+ AI modelsMCP + RESTR2 storage included
Read the docs →

How It Works

Three API calls. Full creative pipeline.

01

Register Your Bot

POST /api/v1/auth/bot-register with your bot name. Get back an api_key and dashboard_token. 30 seconds.

POST /api/v1/auth/bot-register {"bot_name": "my-agent"} → { "api_key": "vyd_...", "dashboard_token": "dt_..." }
02

Create a Job

POST /api/v1/jobs with a workflow and input. Async job queued immediately. Supports video, image, speech, ASMR, lipsync.

POST /api/v1/jobs Authorization: Bearer vyd_... { "workflow": "generate_video", "input": { "prompt": "Drone shot at sunset", "model": "veo3" } } → {"id": "abc-123", "status": "running"}
03

Get Results

Poll /api/v1/jobs/{id} or receive webhook. R2 public URL returned. Or list finished assets via /api/v1/assets.

GET /api/v1/jobs/abc-123 → { "status": "completed", "result": { "videoUrl": "https://pub-xxx .r2.dev/video/..." } }

Native OpenClaw Media Provider

Vydra is built into OpenClaw.

OpenClaw ships Vydra as the bundled vydra provider for image, video, and speech generation. Add one API key, pick Vydra models as defaults, and your agents can create media without custom API glue.

Imagesimage_generate

vydra/grok-imagine

Videovideo_generate

vydra/veo3 or vydra/kling

Speechtts

vydra/elevenlabs/tts

OpenClaw setup docs

OpenClaw config

# Add your Vydra key to OpenClaw
export VYDRA_API_KEY="vydra_live_..."
openclaw onboard --auth-choice vydra-api-key
# Select Vydra as the default media provider
"agents": {
"defaults": {
"imageGenerationModel": {
"primary": "vydra/grok-imagine"
},
"videoGenerationModel": {
"primary": "vydra/veo3"
}
}
}
# Then call Vydra from any OpenClaw agent
image_generate(model="vydra/grok-imagine")
video_generate(model="vydra/veo3")

Capabilities

Everything your agent needs.

Full creative pipeline accessible via REST API or MCP protocol.

ENDPOINT

Bot Registration

One POST to /api/v1/auth/bot-register. Get API key + dashboard token. No human signup needed.

FEATURE

OpenClaw + MCP

Use Vydra as an OpenClaw media provider or install vydra-mcp via npx. 13 agent tools plus first-class image, video, and speech generation.

FEATURE

12+ AI Models

Gemini, Grok, Kling, Veo3, Flux, NanoBanana. All through one unified REST API.

FEATURE

Credit System

Pay per generation. Purchase credits via API or dashboard. Plans available for higher limits.

ENDPOINT

Async Job Queue

POST /api/v1/jobs to submit. Poll for status or use webhooks. List assets when ready.

FEATURE

Image + Video

Generate images, edit them, generate video, lip-sync, motion transfer. Full creative pipeline.

ENDPOINT

Webhooks

Get POST notifications when jobs complete. Zero polling needed for production workflows.

FEATURE

Human Dashboard

Your client gets a branded portal to manage credits, view history, and monitor usage.

Use Cases

What agents build with Vydra.

Social Content Pipeline

OpenClaw agents can call Vydra through the bundled media provider to generate images, videos, and speech inside the same automation loop.

Ad Creative at Scale

Test 50 video variations without a production team. Agents generate, analyze performance, iterate.

Product Demos

Agents generate fresh product walkthroughs from a changelog. Video docs that stay current.

Multi-Agent Pipelines

One agent researches. One writes. One generates. One publishes. Vydra is the creative layer they all share.

See the API →

Integration

Three calls. That's it.

Register Bot

// POST /api/v1/auth/bot-register
{
"bot_name": "my-video-agent"
}
// Response
{
"api_key": "vyd_sk_...",
"dashboard_token": "vyd_dt_...",
"credits": 100
}

Generate Video

// POST /api/v1/jobs
{
"workflow": "generate_video",
"input": {
"prompt": "A cinematic drone shot over mountains",
"model": "veo3"
}
}
// GET /api/v1/jobs/{job_id}
{
"status": "completed",
"result": { "videoUrl": "https://pub-xxx.r2.dev/video/..." },
"creditsCharged": 175
}

Pricing

Simple, usage-based pricing.

Start free. Scale when you need to.

Free

$0forever
  • 100 API credits
  • Bot registration
  • All 12+ models
  • Community rate limits

Starter

$7/month
  • 2,000 API credits/mo
  • All models
  • Standard rate limits
  • Email support

Most Popular

Creator

$19/month
  • 5,500 API credits/mo
  • All models
  • Priority rate limits
  • Priority support
  • MCP + REST API access

Pro

$49/month
  • 14,000 API credits/mo
  • All models
  • Highest rate limits
  • Priority support
  • Full MCP + REST + webhooks