Gamma AI Image Generator - Bulk Generate Images via Gamma (Flux 2, Nano Banana, Imagen 4, GPT Image 2)

Bulk-generate AI images through your Gamma.app account at scale. Access 50+ image models including Flux 2 Pro, Flux 2 Max, Nano Banana 2, Nano Banana Pro HD, GPT Image 2, Imagen 4 Ultra, Ideogram 3.0, Recraft V4, and dozens more. Process 100 prompts in a single run, attach up to 4 style reference images, and export image URLs in JSON, CSV, Excel, or via webhook - no Gamma API key needed, just your session cookie.

What is this? A wrapper around Gamma's internal image generation API. Your existing Gamma subscription's credits power every generation - you pay Apify only for the orchestration. Cheapest way to drive Gamma at scale from automation pipelines.

What this Gamma image scraper / generator does

Reverse-engineered against Gamma's web app, the actor:

• Accepts a list of up to 100 prompts per run.
• Lets you pick from all 50 image models Gamma's API accepts - including 9 NEW 2026 models (Flux 2 family, Nano Banana 2, Recraft V4, GPT Image 2) that aren't documented anywhere else.
• Supports up to 4 style reference images per run - any public HTTP URL works; the actor handles the Transloadit upload pipeline behind the scenes.
• Returns clean dataset rows with the generated image URL, dimensions, prompt, model, and metadata.
• Built-in multi-prompt concurrency, retry with backoff, and request delays to keep Gamma happy.

Perfect for bulk marketing asset generation, content pipelines, A/B style testing, AI batch image workflows, and automation projects that need Gamma's image quality at scale.

Why use this Gamma actor

• 50+ image models accessible - the full set the Gamma API will accept, including new releases the official UI hasn't fully exposed yet.
• Bulk prompts - 100 prompts × 3 images = 300 images per run.
• Style reference uploads handled automatically - paste any image URL, the actor uploads it into your Gamma workspace.
• No Gamma API key required - works with your session cookie.
• Use your own subscription's credits - no per-image markup from Apify on the underlying generation cost.
• Retry + concurrency built in - production-ready for automation pipelines.

What data you get per image

{
  "url": "https://cdn.gamma.app/{workspaceId}/generated-images/{id}.png",
  "width": 1440,
  "height": 1440,
  "prompt": "your prompt here",
  "model": "flux-2-pro",
  "aspectRatio": "square",
  "artStylePreset": "photorealistic",
  "styleReferenceCount": 0,
  "createdAt": "2026-05-20T12:02:47.070Z",
  "id": "0fl16q66wx04khr"
}

How to use the Gamma AI Image Generator

1. Click Try for free on the actor page.
2. Log in to Gamma in your browser, open DevTools → Network → click any API request → copy the entire Cookie request header → paste into cookie.
3. From the same DevTools, find your workspace ID in any GraphQL request body (search for workspaceId) → paste into workspaceId.
4. Enter your prompts in the prompts array (one prompt per line/item).
5. Pick a model (or enable Model Auto-Select) and set Image Count, Aspect Ratio, Art Style.
6. Optional: paste up to 4 image URLs into Style Reference Images for style transfer.
7. Click Save & Start. Download dataset in JSON, CSV, Excel, or stream via API.

Top use cases

• Bulk marketing assets - generate 100s of social ad creatives per campaign.
• Content pipelines - feed prompts from your CMS, get back image URLs to embed.
• AI batch image workflows - drive Gamma from Make/Zapier/n8n via webhook.
• A/B style testing - run the same prompt across 5 art-style presets to compare.
• Style transfer projects - attach reference images to anchor the brand look.
• AI design experimentation - try the same prompt across Flux 2, Nano Banana, GPT Image 2 to find the best model.
• Game / product art - bulk generate concept art variants from a list of prompts.

Models supported (50)

The actor exposes every model Gamma's API will accept. Models marked ✿ref support style reference images.

Tier	Models
Recommended	Flux 2 Pro, GPT Image 2
Basic / Fast	Flux 2 Fast, Flux Fast 1.1, Flux Schnell, Ideogram 3.0 Flash, Imagen 3 Fast, Luma Photon Flash, Flux Kontext Fast
Advanced	Flux 1 Pro, GPT Image, GPT Image Mini, GPT Image 2 Mini, Ideogram 3.0 Turbo, Imagen 4 Fast, Leonardo Phoenix, Luma Photon, Nano Banana 2 Mini ✿ref, Recraft V4
Premium	Flux 2 Max ✿ref, Flux Kontext Pro, Flux 2 Klein, GPT Image Detailed ✿ref, Ideogram 3.0 ✿ref, Imagen 4, Nano Banana ✿ref, Nano Banana Pro ✿ref, Nano Banana 2 ✿ref, Recraft V3, Recraft V3 Vector, Recraft V4 Vector
Ultra	Flux Ultra, Flux Kontext Max, Ideogram 3.0 Quality ✿ref, Imagen 4 Ultra, Nano Banana Pro HD ✿ref, Nano Banana 2 HD ✿ref, Recraft V4 Pro
Legacy	DALL-E 3, Imagen 3, Playground 3, Ideogram 2.0 family, Ideogram Gamma Custom, Leonardo SDXL Gamma, Leonardo SDXL MrPeepers, several GPT Image legacy variants

If you attach style reference images and your selected model does not support them, the run auto-switches to Model Auto-Select so the generation still works.

Art style presets (6)

photorealistic, illustration, abstract, 3D, lineArt, custom (provide your own Art Style Prompt).

Aspect ratios (3)

square (1:1), landscape (16:9 / 4:3), portrait (9:16 / 3:4 / 4:5).

Integrations

Apify API

$curl "https://api.apify.com/v2/datasets/{DATASET_ID}/items?format=json"

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")
run = client.actor("goat255/gamma-unlimited-ai-images").call(run_input={
    "cookie": "gamma_session=...",
    "workspaceId": "your_workspace_id",
    "prompts": ["a serene mountain lake at sunrise", "neon city skyline"],
    "model": "flux-2-pro",
    "count": 2,
    "aspectRatio": "landscape",
    "artStylePreset": "photorealistic",
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["url"])

JavaScript / Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_API_TOKEN' });
const run = await client.actor('goat255/gamma-unlimited-ai-images').call({
    cookie: 'gamma_session=...',
    workspaceId: 'your_workspace_id',
    prompts: ['cyberpunk diner at night'],
    model: 'gpt-image-2',
    count: 1,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach(i => console.log(i.url));

No-code

Push results to Google Sheets, Slack, Zapier, Make, Amazon S3, HubSpot, or any webhook via Apify Integrations.

FAQ

Do I need a Gamma API key?

No. The actor works with your existing Gamma session cookie. Gamma does not currently offer a public image-generation API.

Where do I find my cookie and workspace ID?

Open Gamma in your browser → DevTools → Network tab → click any request to api.gamma.app or ai.api.gamma.app → look at the request headers. The full Cookie: value goes into the cookie input. The workspaceId is in any GraphQL request body (search the request body for "workspaceId").

Does this consume my Gamma credits?

Yes. Each image is billed to your Gamma subscription at Gamma's listed credit rate per model. The actor only orchestrates the calls; it does not bypass Gamma's billing.

Can I attach reference images from any URL?

Yes - any public HTTP image URL works. The actor downloads it, uploads it into your Gamma workspace via Transloadit, then uses the resulting CDN URL as a style reference. Max 4 references per run.

Which models support style references?

12 of the 50 image models, marked ✿ref in the model list above (the Nano Banana family, Ideogram 3.0 / Quality, Flux 2 Max, GPT Image / GPT Image Detailed / GPT Image 2). When references are attached, the actor auto-switches to a reference-capable model if needed.

How many prompts can I run at once?

Up to 100 prompts per run × up to 3 images per prompt = up to 300 images. Concurrency is configurable (default 3 parallel prompts).

What about cookies expiring?

Gamma sessions are long-lived but do expire. If a run fails with an auth error, log back into Gamma and copy the fresh Cookie header.

Can I export to CSV / Google Sheets / Excel?

Yes - the dataset supports JSON / CSV / Excel / XML / HTML export and direct integrations.

Is this rate-limited?

The actor uses retries with exponential backoff. The default 1.5s delay between requests keeps you well under Gamma's rate ceiling for normal use.

Pricing

Pay-per-event on Apify. See the actor's Pricing tab on the Store. Cheap orchestration; Gamma's own credits handle the underlying generation cost.

Related Apify actors

• 📷 Instagram Posts Scraper - bulk export Instagram posts.
• 👤 Instagram Profile Scraper - Instagram profiles with emails.
• 📍 Google Maps Scraper - local business leads.

Support

Open an issue on the actor page if a model stops working or you'd like more options exposed. Reviews are read and replied to.