SuperAPI AI Directory
Partner中文AI newsFavorites
Skillsapiyi-gpt-image-2-gen

apiyi-gpt-image-2-gen

Coding & development

View repository

Updated 10-16 01:30

About this skill

Image generation and editing (GPT Image 2 official release)

Image generation skill based on the GPT Image 2 model (gpt-image-2) official release on the API易 platform, which can help users generate images through natural language, accessed via API易's domestic proxy service, and supports both Node.js and Python runtime environments. gpt-image-2 is the official release GPT image generation model on the API易 platform, supporting precise size/quality control (including 4K), billed by token.

Usage guide

Follow these steps:

Step 1: Analyze requirements and extract parameters

  1. Clarify intent: distinguish whether the user needs [text-to-image] (generate a new image), [image-to-image] (edit/modify an existing image), or [multi-image fusion].
  2. Prompt analysis：
    • Use the user's original complete input: use the user's original complete problem/requirement description (verbatim) directly as the main body of the -p prompt, avoiding rewriting, summarizing, or secondary creation on your own, to prevent loss of details.
    • Confirm first when supplementation is needed: if information is insufficient (for example, missing style, number of subjects, camera language, scene details, text content, prohibited elements, etc.), first ask the user to confirm; after the user confirms, then append the supplementary contentin an "append" mannerafter the original prompt.
    • Example:
      • User input: "Help me generate an image of a cat, the style should be a bit cute."
      • Positive example: directly use the user input as the prompt:-p "帮我生成一张猫的图片，风格要可爱一点。"
      • Negative example: arbitrarily rewriting it as "generate an image of a cute-style cat" will lose the details and tone of the user's original input.
      • If details need to be supplemented (for example, color, background, etc.), first ask and confirm: "What color do you want the cat to be? Any requirements for the background?" After the user answers, then append to the prompt:-p "帮我生成一张猫的图片，风格要可爱一点。猫是橘色的，背景是草地。"
  1. Organize key parameters：
    • Prompt (required): the final prompt after prompt analysis (default = the user's original complete and consistent input; supplementary information is appended only after user confirmation).
    • Filename (optional): output image filename/path (must include a random identifier to avoid duplication). If not provided, the script will automatically generate a filename with a timestamp. It is recommended to generate a reasonable filename based on the content (for example cat_in_garden.png), and avoid using generic names.
    • Size (optional): output size.
      • Preset values:1024x1024、1536x1024、1024x1536、2048x2048、2048x1152、3840x2160、2160x3840
      • Custom sizes can also be used (satisfying: longest side ≤3840, both sides multiples of 16, ratio ≤3:1, total pixels 0.65–8.3MP)
      • Defaults to model adaptive (auto)
    • Quality (optional): quality level.low(sketch/batch),medium(daily),high(final draft/fine text),auto(default)
    • Output Format (optional)：png(default),jpeg、webp
    • Output Compression (optional): output compression rate (0-100), only effective for jpeg/webp
    • Note: this model uses the official stable endpoint, unlike the unofficial reverse-engineered gpt-image-2-all.

Step 2: Environment Check and Command Execution

  1. Check the environment: confirm whether the APIYI_API_KEY environment variable is set (usually assumed to be set; if execution fails, then prompt the user)���

  2. Build and run the command：

    • Try the Node.js version first: if Node is available in the environment (thenode command is available), prefer scripts/generate_image.js(zero dependencies, parameters consistent with Python).
    • If Node is unavailable, use the Python version: use scripts/generate_image.py。

    Text-to-image command template (prefer Node.js):

    node scripts/generate_image.js -p "{prompt}" -f "{filename}" [-s {size}] [-q {quality}] [-o {output_format}]

    Image-to-image command template (prefer Node.js):

    node scripts/generate_image.js -p "{edit_instruction}" -i "{input_path}" -f "{output_filename}" [-s {size}] [-q {quality}]

    Multi-image fusion command template (prefer Node.js):

    node scripts/generate_image.js -p "融合图1和图2的风格" -i ref1.png ref2.png -f "merged.png" [-s {size}] [-q {quality}]

    (optional) Python version command template (when Node is unavailable)：

    python scripts/generate_image.py -p "{prompt}" -f "{filename}" [-s {size}] [-q {quality}] [-o {output_format}]
python scripts/generate_image.py -p "{edit_instruction}" -i "{input_path}" -f "{output_filename}" [-s {size}] [-q {quality}]

⏱️ Long-running task handling strategy

1. Pre-task notice

Must inform the user before execution：

  • "Image generation has started, it is expected to take 120-150 seconds, please wait patiently"

2. 🎨 Best practice example

"Image is being generated, expected to complete in 120-150 seconds...\n⏳ Generating...\n(high + 2K/4K complex scenes may take longer, please wait patiently)"

Step 3: Result Feedback

  1. Execution feedback: wait for the terminal command to finish executing.
  2. Success: inform the user that the image has been generated and indicate the save path.
  3. Failure：
    • If it indicates the API Key is missing, guide the user to set the environment variable.
    • If it indicates a network error, suggest the user check the network or try again later.

Command-line usage examples

Generate a new image

python scripts/generate_image.py -p "图片描述文本" -f "output.png" [-s {size}] [-q {quality}] [-o {output_format}]

Example:

# 基础生成
python scripts/generate_image.py -p "一只可爱的橘猫在草地上玩耍" -f "cat.png"

# 指定尺寸和画质
python scripts/generate_image.py -p "日落山脉风景" -f "sunset.png" -s "2048x1152" -q "high"

# 竖版高清图片（适合手机壁纸）
python scripts/generate_image.py -p "城市夜景" -f "city.png" -s "2160x3840" -q "high"

# 输出为JPEG
python scripts/generate_image.py -p "风景照片" -f "landscape.jpg" -s "3840x2160" -q "high" -o "jpeg"

(optional) Node.js version example:

# 基础生成
node scripts/generate_image.js -p "一只可爱的橘猫在草地上玩耍" -f "cat.png"

# 指定尺寸和画质
node scripts/generate_image.js -p "日落山脉风景" -f "sunset.png" -s "2048x1152" -q "high"

Edit an existing image

python scripts/generate_image.py -p "编辑指令" -f "output.png" -i "path/to/input.png" [-s {size}] [-q {quality}]

Example:

# 修改风格
python scripts/generate_image.py -p "将图片转换成水彩画风格" -f "watercolor.png" -i "original.png"

# 添加元素
python scripts/generate_image.py -p "在天空添加彩虹" -f "rainbow.png" -i "landscape.png" -q "high"

# 替换背景
python scripts/generate_image.py -p "将背景换成海滩" -f "beach-bg.png" -i "portrait.png" -s "2048x2048"

(optional) Node.js version example:

# 修改风格
node scripts/generate_image.js -p "将图片转换成水彩画风格" -f "watercolor.png" -i "original.png"

# 多图参考图融合（最多5张）
node scripts/generate_image.js -p "把图1的人物放进图2的场景" -i ref1.png ref2.png -f "merged.png"

Command-line parameter description

The parameters of the Python and Node.js versions are consistent (short parameters and long parameters are equivalent).

Parameter Required Description
-p / --prompt Yes Image description (text-to-image) or editing instruction (image-to-image). Preserve the user's original complete input.
-f / --filename No Output image path/filename; if not provided, a filename with a timestamp is automatically generated.
-s / --size No Output size: 1024x1024 / 1536x1024 / 1024x1536 / 2048x2048 / 2048x1152 / 3840x2160 / 2160x3840 or a custom size.
-q / --quality No Quality level: low / medium / high / auto (default auto).
-o / --output-format No Output format: png (default) / jpeg / webp.
-c / --output-compression No Output compression rate (0-100), only effective for jpeg/webp.
-i / --input-image No Input image path for image-to-image; multiple images can be passed (up to 5). Passing this parameter enters edit mode.

File Resource Description

Resource Description
scripts/generate_image.js Node.js version (zero dependencies, preferred)
scripts/generate_image.py Python version (alternative)
references/size-guide.md Size and aspect ratio control documentation, use when needed, load on demand
references/batch-template.md Batch generation configuration template, use when batch generation is needed, load on demand

Batch generation

When the user needs to generate multiple images at once (batch generation):

  1. Load configuration template：references/batch-template.md — includes JSON configuration format description and usage examples
  2. Obtain/generate JSON file: the user can provide a JSON file themselves, or describe the requirements and the AI will generate it based on the requirements
  3. Execute one by one: read the prompts array, execute the generation command one by one, and report the result after each image is completed
  4. Summary feedback: after completion, inform the number of successes and the list of image paths

Note: total batch task time = single image time (120-150 seconds) × number of images, please inform the user of the estimated duration in advance.

Quality Description

Quality Description Applicable scenarios
low Sketch/batch generation Quick preview, multiple iterations
medium Daily General use
high Final draft/fine text Final output, images containing text
auto Default Determined by the model

Output Format Description

Format Description Applicable scenarios
png No compression, transparent background Requires transparent background, retains best quality
jpeg Compressed Photos, storage space sensitive
webp Modern format Web use, balances quality and size

Note: the b64_json field is pure base64, without the data:image/...;base64, prefix. The client needs to:

  • Write file:base64.b64decode(b64_str) → write to disk
  • Browser rendering: prepend the prefix yourself data:image/png;base64, + b64

Precautions

  • The API key must be set, and can be provided via environment variable or command line parameter
  • Image generation time: about 120-150 seconds; high + 2K/4K complex scenes may take longer
  • When editing images, use multipart/form-data to upload the reference image
  • Ensure the output directory has write permission
  • Billed by token (not by image)

API Key Setting and Acquisition

How to get the API Key

If you do not have an API key yet, please go to https://api.apiyi.com to register an account and apply for an API Key.

Steps to obtain:

  1. Visit https://api.apiyi.com
  2. Register/log in to your account
  3. Create an API key in the console
  4. Copy the key and set the environment variable or use it in the command line

Set the API Key

The script obtains the API key from the environment variable APIYI_API_KEY .

Set the environment variable:

# Linux/Mac
export APIYI_API_KEY="your-api-key-here"

我的电脑高级设置中设置环境变量或者执行set 命令：
# Windows CMD
set APIYI_API_KEY=your-api-key-here

# Windows PowerShell
$env:APIYI_API_KEY="your-api-key-here"

API Endpoint Description

Text-to-image endpoint: POST /v1/images/generations

Text-to-image endpoint, uses JSON format for requests.

Image-to-image endpoint: POST /v1/images/edits

Image-to-image endpoint, requested using multipart/form-data format. Upload reference images (up to 5) + instructions for single-image editing and multi-image fusion.

The order of reference images matters; in the prompt, you can use "image 1/image 2/image 3" to refer to them.

Model information

  • Model name: gpt-image-2
  • Image generation speed: about 120-150 seconds (complex 4K scenes may take longer)
  • Output resolution: 1024x1024 / 1536x1024 / 1024x1536 / 2048x2048 / 2048x1152 / 3840x2160 / 2160x3840 or custom
  • Default response format: b64_json (pure base64, no prefix)
  • Quality level: low / medium / high / auto
  • Output format: png / jpeg / webp
  • Supported capabilities: text-to-image, single-image editing, multi-image fusion
  • Billing method: billed by token

gpt-image-2 (official relay) vs gpt-image-2-all (official reverse) comparison

Feature gpt-image-2 gpt-image-2-all
Nature Official standard version Official reverse version
Billing By token Flat $0.03/image
Endpoint /v1/images/generations, /v1/images/edits /v1/chat/completions
Upload reference images multipart form-data base64 data URL
Download images b64_json (pure base64) url or b64_json (with prefix)
Multi-image fusion image[] array up to 5 images chat multiple image_url
Size control Explicit size parameter prompt description
Speed About 120-150 seconds About 60-300 seconds

Author introduction

  • The love for One Piece is everywhere
  • My WeChat official account: 无处不在的技术