# FramePack Studio GPT Image 2.5

> Canonical: https://framepack.studio/ko/gpt-image-2-5

Keep the subject separated from the background and specify where it sits in the frame. Review hands, edges and objects before saving a source image for animation, since later motion can amplify defects already present.

## FramePack Studio GPT Image 2.5 opening frames

### Why check a still before animation?

Inspect fingers, outlines and overlapping objects at the source stage. A malformed edge can become more distracting when the image is used in an animation workflow.

## Technical specs

### 20,000

Prompt characters

### 30 MB

Per reference file

### 1K–4K

Output size options

### 1–4

Images per batch

## Best for animation source preparation

### A practical starting point

Animators can establish the opening composition and identify image defects while they are still static and easy to inspect.

## Questions about opening frames

### Can I revise an opening frame?

Load the frame in image mode and identify the intended change. References may be JPEG, PNG or WebP; each file must fit the 30 MB upload limit.

### How do I choose dimensions?

Pick a ratio first and then a resolution. The 16:27, 27:16, 8:9 and 9:8 options allow only 1K, while the other crops permit 2K and 4K.

### Can I separate foreground elements?

A transparent background can help prepare a foreground asset. High-resolution Sunburst requires a prompt without setting or shadow; explicitly request extraction from a reference, then verify its edges.

## Continue with your image

## 관련 도구

### [FramePack Studio 이미지 생성기](https://framepack.studio/ko/ai-image-generator)

### [FramePack Studio 음악 생성기](https://framepack.studio/ko/ai-music-generator)

### [FramePack Studio 비디오 생성기](https://framepack.studio/ko/ai-video-generator)

_최종 업데이트: 2026년 7월 21일_