Back to Blog
๐Ÿ–ผ๏ธ Image Processing6 min readJun 2, 2026

Generate and Edit Images at Scale with AI Batch Processing

From generating product shots to removing backgrounds in bulk โ€” learn how AI Batch Processor handles image generation and editing the same way it handles text: one instruction, hundreds of images.

Why Image Batch Processing?

If you have ever needed to generate 50 product photos, remove backgrounds from 200 images, or create multiple variations of a design, you know how tedious manual editing can be. Image batch processing applies the same idea as text batch processing โ€” you define one instruction and let AI handle the rest, at scale.

AI Batch Processor now supports five image modes that cover the most common visual content needs:

The Five Image Modes

1. Free Generate โ€” Text-to-Image from Scratch

Each line of input is a complete prompt. The AI generates a unique image for every line, making it ideal for concept exploration, mood boards, or visual brainstorming.

Example Input:

A minimalist coffee cup on a marble countertop, morning light
Same coffee cup held by a person in a cozy office setting
Coffee cup with steam, dark background, product photography style

Cost: 20 credits per image. Each line produces one image independently.

2. Reference Generate โ€” Consistent Visuals with Variations

Upload one reference image, then provide batch content lines describing what changes you want. The AI keeps the visual style consistent while varying the content. This is perfect for e-commerce product shots where you need the same product shown from different angles or in different settings.

How It Works:

Reference Image: [product photo] + Shared Instruction: "Show the product in a lifestyle setting"
Batch Content 1: "On a kitchen counter with fresh ingredients"
Batch Content 2: "Being held by a person in a park"
Batch Content 3: "On a wooden table with coffee beans"

Cost: 30 credits per image.

3. Replace โ€” Swap Objects in Images

Upload an image and tell the AI what to replace. For example, replace a plain background with a scenic view, or swap an old product version with a new one. Each item describes the target and what to replace it with.

4. Remove โ€” Take Elements Out Cleanly

Remove unwanted objects, watermarks, text overlays, or people from images. The AI fills in the background naturally, making it look like the element was never there. Works well for cleaning up product photos or preparing assets for design use.

5. Remake โ€” Reimagine with a Different Vibe (Coming Soon as Guided Mode)

Pure Remake generates a fresh version of the image from its description. Directed Remake lets you specify what to keep and what to change โ€” like "keep the product, change the background to night time."

๐Ÿ’ก Note: Replace, Remove, and Remake use image-to-image generation. The AI edits your uploaded image rather than creating one from scratch, which means the output preserves the original composition and quality.

How Credits Work for Images

Image processing consumes credits differently from text processing because generating images requires significantly more compute resources:

  • Free Generate: 20 credits per image
  • Reference Generate / Replace / Remove / Remake: 30 credits per image
  • Credit packs can be used for any image mode
  • Failed image generations do not consume credits

Your monthly quota applies to both text and image credits. If your plan includes 500 credits per month, you can mix text and image processing freely within that limit.

Image Batch Limit

Each plan has a maximum number of images you can process in a single batch:

  • Free: up to 1 image per batch
  • Starter: up to 5 images per batch
  • Pro: up to 15 images per batch
  • Business: up to 30 images per batch
  • Enterprise: up to 50 images per batch

Credit packs also increase your batch limit (Small pack grants Starter-level limit, Medium grants Pro-level, Large grants Business-level).

Technical Details

Behind the scenes, image generation uses the Doubao Seedream model via Volcengine Ark. Each image supports resolutions from 1K to 4K, and you can choose between single-image or auto-sequential generation for multiple variations.

Upload limits: maximum 10MB per image, 60MB total per batch. Supported formats are PNG, JPG, and WebP.

When to Use Image Batch vs Text Batch

Both tools share the same workspace layout and quota system, but they serve different needs:

  • Text Batch โ€” process structured text: translate, rewrite, extract, classify, generate
  • Image Batch โ€” generate and edit visuals: product photos, concept art, background removal, variations

You can use them side by side. Many users start with Image Batch to create visuals, then switch to Text Batch to generate descriptions or SEO metadata for those images.

Ready to Try Image Batch?

No setup required. Sign up for free, navigate to the Image tab, and start generating. Your first 20 monthly credits cover one free-generation image to test the quality.

Go to Image Batch
Share this article