omniflash-sdk

Node.js / TypeScript SDK for Gemini Omni Flash — generate short video clips (with synchronized audio) and images using Google's Gemini Omni Flash family of models.

Gemini Omni Flash wraps the Omni Flash family (seedance-2 for text/image → video + audio, gpt-image-2 and nano-banana-2 for text/image → image) behind one simple REST API. This SDK is a thin, dependency-free wrapper around the global fetch.

Install

npm i omniflash-sdk
# or: pnpm add omniflash-sdk / yarn add omniflash-sdk

Get an API key

Sign in at Gemini Omni Flash, open the account page, then create a sk-… token.

export OMNIFLASH_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Quick start

import { OmniFlash } from "omniflash-sdk";

const client = new OmniFlash(); // reads OMNIFLASH_API_KEY

const task = await client.run({
  model_id: "seedance-2",
  prompt: "a kettle whistles as steam rises, cozy kitchen, warm light",
  aspect_ratio: "16:9",
});

console.log(task.video_url);
console.log(task.audio_url); // synchronized audio track

Lower level: create + poll

import { OmniFlash, TaskStatus } from "omniflash-sdk";

const client = new OmniFlash({ apiKey: "sk-..." });

const created = await client.createTask({
  model_id: "gpt-image-2",
  prompt: "cyberpunk corgi, neon rim light",
  aspect_ratio: "1:1",
});

let task = created;
while (task.task_status === TaskStatus.Queued || task.task_status === TaskStatus.Running) {
  await new Promise((r) => setTimeout(r, 3000));
  task = await client.getTask(task.task_id);
}
console.log(task.image_url);

Models

| model_id | Modality | Output | | --------------- | ------------------ | ----------------------- | | seedance-2 | text/image → video | video_url + audio_url | | gpt-image-2 | text/image → image | image_url | | nano-banana-2 | text/image → image | image_url |

See current models and pricing on Gemini Omni Flash.

API

• new OmniFlash({ apiKey?, baseUrl?, fetch?, timeoutMs? })
• client.createTask({ model_id, prompt, image_urls?, aspect_ratio? }) → Promise<Task>
• client.getTask(taskId) → Promise<Task>
• client.run({ ...createTaskInput, pollIntervalMs?, maxWaitMs?, signal? }) → Promise<Task>
• Task: { task_id, task_status (1=queued, 2=running, 3=success, 4=failed), image_url?, video_url?, audio_url?, request_id?, task_type?, model_id?, credits?, created_at? }

Errors throw an OmniFlashError with optional code (business) and status (HTTP).

Links

• Website & account: Gemini Omni Flash
• API docs: https://omniflash.net/api-docs

License

MIT