gpt-image系列图片生成

1. 功能介绍

根据文本提示生成一张或多张高质量图像，支持：

GPT-Image 系列模型（gpt-image-1.5 / gpt-image-1）

多尺寸、多质量、背景透明、流式输出等精细控制

自带内容审核，可自定义审核严格度

异步 / 同步双模式，适配不同业务场景

2. 请求参数

字段	类型	必填	说明	备注
prompt	string	✅	描述文本	长度 ≤ 32 k（gpt-image-1.5）
model	string	✅	模型	可选：gpt-image-1.5 / gpt-image-1
n	int		生成张数	1-10；dall-e-3 仅支持 1
size	string		尺寸	1024x1024、1536x1024（横向）、1024x1536（纵向）、auto ，尺寸选择 `auto` 时，模型会根据提示词自动给出最佳比例。
quality	string		质量	low / medium / high / auto
background	string		背景	transparent / opaque / auto
output_format	string		输出格式	png / jpeg / webp
output_compression	int		压缩率	0-100；仅 jpeg/webp 生效
moderation	string		审核严格度	low（宽松） / auto
stream	bool		流式输出	仅 GPT-Image 系列支持
partial_images	int		流式分片数	0-3；0=仅返回最终图
async	query		是否异步	传 `?async=true` 时返回 task_id

3. 注意事项

✅ 请根据选择的模型严格遵守 prompt 字符长度限制

📏 不同模型支持的尺寸、质量参数不同，需提前确认兼容性

🔒 敏感内容将被系统拦截，moderation参数建议保持默认auto模式

📊 价格按生成图像的Tokens计算，不同模型/质量的Token消耗不同（详见定价部分）

🔄 可通过任务查询接口主动检查异步任务的生成进度

4. 价格

模型	文字输入	文字输出	图片输出	备注
gpt-image-1	5 PTC/1M Tokens		40 PTC/1M Tokens	高品质模型
gpt-image-1.5	5 PTC/1M Tokens	10 PTC/1M Tokens	32 PTC/1M Tokens	平衡型模型

最终价格以请求返回的Tokens消耗数量为准

gpt-image-1: 价格参考

Request

Authorization

Provide your bearer token in the

Authorization

header when making requests to protected resources.

Example:

Authorization: Bearer ********************

Body Params application/jsonRequired

Example

{
    "prompt": "string",
    "model": "gpt-image-1",
    "size": "1024x1024",
    "n": 1,
    "background": "transparent",
    "moderation": "low",
    "quality": "auto",
    "output_format": "png",
    "output_compression": 100,
    "partial_images": "0",
    "stream": false
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

Responses

🟢200OK

application/json

Body

Example

{
    "data": [
        {
            "b64_json": "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

gpt-image系列图片生成

1. 功能介绍#

2. 请求参数#

3. 注意事项#