GPT Image 2.5 Sunburst
OpenAI 面向高质量生成与精确编辑的最强图片模型。
产品定位
最高图片能力
适合复杂生成与精确编辑
输入类型
文本 + 图片
生成或编辑时最多使用 14 张参考图
质量档位
6 档
从 low 到 max，并支持自动选择
执行方式
同步 + 异步
可等待直接响应，或通过轮询和回调获取结果
gpt-image-2.5-sunburst试玩区
gpt-image-2.5-sunburstClick or drag
分辨率 (sent to backend)
Aspect Ratio (sent to backend)
Quality (sent to backend)
Output format (sent to backend)
暂无输出
定价
| 用量指标 | Standard 价格 |
|---|---|
| 文本输入 | $4590$5 / 100 万 Token10% off |
| 图片输入 | $7290$8 / 100 万 Token10% off |
| 图片输出 | $27090$30 / 100 万 Token10% off |
Text input · $4.5 · 1M tokens
Image input · $7.2 · 1M tokens
Image output · $27 · 1M tokens
README
- GPT Image 2.5 Sunburst 是 OpenAI 最强的图片生成与编辑模型，适合编辑精度和整体结果质量比最低延迟更重要的复杂工作流。
- 模型支持基于文本和图片输入进行生成与编辑。ReachAPI 支持最多 14 张 HTTPS 参考图、1K、2K 和 4K 分辨率档位、三种标准比例，以及 PNG、JPEG 或 WebP 输出。
- 质量可设置为 low、medium、high、xhigh、max 或 auto。虽然 Sunburst 与 Flare 使用相同的 Token 单价，但不同质量和图片尺寸的 Token 消耗可能不同。
- 该模型目前在 ReachAPI 处于内部预览阶段。可通过 OpenAI 兼容的同步接口直接等待结果，也可通过异步任务接口使用 URL 参考图、任务轮询和可选的终态回调。
API
API
端点
同步生成
POST https://direct.reachapi.ai/v1/images/generations同步编辑
POST https://direct.reachapi.ai/v1/images/edits异步任务
POST https://direct.reachapi.ai/v1/images/create认证
Authorization: Bearer YOUR_REACH_API_KEY
示例
同步生成
curl -X POST "https://direct.reachapi.ai/v1/images/generations" \
-H "Authorization: Bearer YOUR_REACH_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.5-sunburst",
"prompt": "A precise editorial product photograph of a silver watch on slate, soft studio lighting",
"size": "1536x1024",
"quality": "high",
"background": "auto",
"moderation": "auto",
"output_format": "png"
}'同步编辑
curl -X POST "https://direct.reachapi.ai/v1/images/edits" \ -H "Authorization: Bearer YOUR_REACH_API_KEY" \ -F "model=gpt-image-2.5-sunburst" \ -F "prompt=Preserve the watch geometry and replace only the background with dark brushed metal" \ -F "image[]=@/path/to/watch.png" \ -F "size=1536x1024" \ -F "quality=xhigh" \ -F "output_format=png"
异步任务
curl -X POST "https://direct.reachapi.ai/v1/images/create" \
-H "Authorization: Bearer YOUR_REACH_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.5-sunburst",
"callback_url": "https://example.com/reachapi/callback",
"input": {
"prompt": "A precise editorial product photograph of a silver watch on slate, soft studio lighting",
"resolution": "2k",
"aspect_ratio": "3:2",
"quality": "high",
"background": "auto",
"moderation": "auto",
"output_format": "png"
}
}'相关模型
GPT-6 Astra
OpenAI's flagship model for demanding end-to-end work.
GPT-5.6 Luna
OpenAI's GPT-5.6 model for efficient, cost-sensitive, high-volume workloads.
GPT-5.6 Sol
OpenAI's GPT-5.6 flagship model for the most demanding production reasoning, coding, and agent workflows.
GPT-5.6 Terra
OpenAI's GPT-5.6 model for strong intelligence with a balanced cost profile.
GPT-5.5
OpenAI's newest frontier model for the most complex coding, reasoning, and professional workflows.
GPT-5.4
OpenAI's frontier model for complex professional, coding, and agentic work at scale.
GPT-5.4 Mini
OpenAI's strongest mini model for coding, computer use, and high-volume agent tasks.
GPT-5.3 Codex
OpenAI's most capable agentic coding model to date.
GPT-5.2
OpenAI's stronger GPT-5-family model for professional, general-purpose, and agentic work.
GPT-5.1
OpenAI's flagship model for coding and agentic tasks with configurable reasoning.
GPT-5.1 Codex
OpenAI's GPT-5.1 variant optimized for agentic coding in Codex workflows.
GPT-5
OpenAI's previous intelligent reasoning model for coding and agentic tasks.
GPT-5 Mini
OpenAI's faster, cost-efficient GPT-5 variant for low-latency and high-volume work.
GPT Image 2.5 Flare
OpenAI's fastest model for high-quality, everyday image generation and editing.
GPT Image 2
OpenAI's state-of-the-art image generation model for high-quality creation and editing with text and image inputs.
GPT Image 2 Async
ReachAPI's asynchronous GPT Image 2 workflow for task-based image generation and editing with callback-ready results.