Quickstart

This guide walks you from zero to a successful image generation, then a video generation. Prerequisites: an account at app.rimp.example and an API key from Settings → API keys.

1. Set your API key

export RIMP_API_KEY="sk_live_..."

$env:RIMP_API_KEY = "sk_live_..."

2. Generate an image (synchronous)

Image models with sub-30s latency are returned directly in the response.

curl -X POST https://api.rimp.example/v1/images \
  -H "Authorization: Bearer $RIMP_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "model": "flux-pro",
    "prompt": "a red rimp on brushed chrome, studio lighting"
  }'

import { RimpClient } from '@rimp/sdk';

const client = new RimpClient({ apiKey: process.env.RIMP_API_KEY! });

const gen = await client.generations.createImage({
  model: 'flux-pro',
  prompt: 'a red rimp on brushed chrome, studio lighting',
});

console.log(gen.outputs[0].url);

from rimp import Client

client = Client(api_key=os.environ["RIMP_API_KEY"])

gen = client.generations.create_image(
    model="flux-pro",
    prompt="a red rimp on brushed chrome, studio lighting",
)

print(gen["outputs"][0]["url"])

Response:

{
  "id": "gen_abc123",
  "object": "generation",
  "status": "succeeded",
  "modality": "image",
  "charged_credits": 40,
  "outputs": [
    {
      "id": "out_xyz",
      "url": "https://cdn.rimp.example/...",
      "mime": "image/png",
      "width": 1024,
      "height": 1024
    }
  ]
}

3. Generate a video (asynchronous)

Video generation returns 202 Accepted immediately. Poll the generation or subscribe to a webhook.

const job = await client.generations.createVideo({
  model: 'veo-3-1-fast',
  prompt: 'slow motion ocean waves crashing on dark volcanic rocks',
  duration_s: 4,
});

// Wait helper polls every 2s, up to 30 minutes
const final = await client.generations.waitFor(job.id);
console.log(final.outputs[0].url);

job = client.generations.create_video(
    model="veo-3-1-fast",
    prompt="slow motion ocean waves crashing on dark volcanic rocks",
    duration_s=4,
)

final = client.generations.wait_for(job["id"])
print(final["outputs"][0]["url"])

4. Compare multiple models in one call

This is Rimp’s headline feature — same prompt, N models, atomic credit reservation.

const cmp = await client.comparisons.create({
  prompt: 'cinematic portrait of a fisherman at golden hour, 35mm film',
  models: ['flux-pro', 'imagen-4', 'ideogram-v2'],
  params_shared: { aspect_ratio: '1:1' },
});

// Poll once, get all three results
const result = await client.comparisons.waitFor(cmp.id);
result.generations.forEach((g) => console.log(g.model, g.outputs[0].url));

If your wallet can’t cover the sum of all three estimates, the API returns 402 Insufficient credits before charging anything.

Next steps

Authentication

Bearer tokens, scoping, rotation, and IP allowlist.

Multi-model deep dive

How routing, capabilities, and pricing work across providers.

Webhooks

Receive generation.completed events instead of polling.

Idempotency

Safely retry without double-charging.

Get started

Core concepts

Webhooks

SDKs

AI tools

1. Set your API key

2. Generate an image (synchronous)

3. Generate a video (asynchronous)

4. Compare multiple models in one call

Next steps

Authentication

Multi-model deep dive

Webhooks

Idempotency

​1. Set your API key

​2. Generate an image (synchronous)

​3. Generate a video (asynchronous)

​4. Compare multiple models in one call

​Next steps

Authentication

Multi-model deep dive

Webhooks

Idempotency

1. Set your API key

2. Generate an image (synchronous)

3. Generate a video (asynchronous)

4. Compare multiple models in one call

Next steps