Introducing /video: Generate video from your agent in one call

Today we're launching /video — the primitive that turns an AI Employee into a video producer. Generate from a prompt, generate from an image, route through the Naïve's model catalog catalog of video models, and hand the resulting URL to /social, /email, or any other primitive that takes media.

The problem: video is the most expensive primitive to skip

For an autonomous content business, video is the most-watched format on every distribution surface — YouTube, TikTok, Instagram Reels, X, LinkedIn. It's also the hardest to produce. The status quo is fragmented: raw model APIs ship clip-length output with no compose surface, DIY pipelines around FFmpeg and TTS take months to assemble, and nothing of the above is a runtime call an Employee can make in one line.

The outcome: most agent-run content businesses post static images and text, leaving the highest-engagement format on the table. Until now.

How /video works

The workflow is straightforward:

1. Pick a model — naive video models returns the current Naïve's model catalog catalog, including Kling, Minimax, Wan, and additional families as they're added.
2. Generate — naive video generate with a prompt (text-to-video) and optional image (image-to-video). Naïve submits the job to fal and returns a job id.
3. Wait / poll — the CLI --wait flag polls for you; the API surface returns the finished URL on the jobs endpoint when the model completes.
4. Compose — hand the URL to /social for distribution, /email for embed, or pair with /image and /brand to keep visual identity consistent across the content stack.

Two ways to produce: CLI or API

1. CLI

naive video generate \
  "60-second explainer of /video in our brand voice" \
  --model fal-ai/kling-video/v3/pro/text-to-video \
  --wait

List available video models with naive video models.

2. API

const response = await fetch("https://api.usenaive.ai/v1/video/generate", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.NAIVE_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "fal-ai/kling-video/v3/pro/text-to-video",
    input: {
      prompt: "60-second explainer of /video in our brand voice",
    },
  }),
});
 
const { job_id } = await response.json();

The job is async — poll with naive video status <job_id> or the jobs endpoint until the video URL is returned.

Image-to-video

For image-to-video workflows — animate a brand still, turn a generated /image into a moving asset — pass an image URL alongside the prompt and pick a model that supports image input:

naive video generate \
  "subtle camera push-in, brand colors" \
  --model fal-ai/kling-video/v3/pro/image-to-video \
  --image-url https://cdn.usenaive.ai/.../still.png \
  --wait

Pair with /image so the same Employee that generates the brand still also drives the motion pass.

What you can build with /video

Produce short-form clips for social distribution — Compose /video with /social so the same Employee that generates the asset also publishes it across the 10 documented social platforms.

Animate brand stills into moving creative — Use /image for the still, /video for the motion, /brand for the palette and tone.

Test creative variants at scale — Batch prompts and models; route the results into /social ad surfaces and feed performance back into the next generation.

Build a research-grounded video producer — Pair /video with /research so the script behind every clip starts from citations the runtime actually fetched.

Get started

Drop this starter prompt into any coding agent to wire up Naïve:

Read https://usenaive.ai/skill.md and use it to set up Naïve in my project.

• Read the docs: usenaive.ai/docs/getting-started/video
• Quickstart: usenaive.ai/docs/getting-started/quickstart
• Background reading: video model catalog
• Join the community on Discord