Create Chat Completion with Image (Non-streaming)

This endpoint is used to send a prompt (containing text or images) to the model, which then returns one or more generated responses. It is specifically designed for models supporting multimodal input (such as image processing, enhancement, and vision-based creation), such as gpt-4o-image-vip.

Endpoint Information

URL: https://api.codingplanx.ai/v1/chat/completions
Method: POST
Content-Type: application/json
Response Type: application/json

Request Headers

Parameter Name	Required	Type	Example Value	Description
Content-Type	Yes	String	`application/json`	Format of the request body.
Accept	Yes	String	`application/json`	Expected response format.
Authorization	No	String	`Bearer {{YOUR_API_KEY}}`	API Key used for authentication.

Request Body

The request body is a JSON object containing the following fields:

Parameter Name	Type	Required	Default	Description
model	String	Yes	-	The ID of the model to use (e.g., `gpt-4o-image-vip`).
messages	Array	Yes	-	A list of messages comprising the conversation. Each object contains `role` (user/assistant/system) and `content`. `content` can be a string or an array of objects containing types and image URLs.
tools	Array	Yes	-	A list of tools the model may call. Currently only functions are supported.
tool_choice	Object	Yes	`none`	Controls which tool is called by the model. `none` means no call, `auto` means automatic selection.
temperature	Number	No	1	Sampling temperature (0-2). Higher values are more random, lower values are more focused.
top_p	Number	No	1	Nucleus sampling. The model considers the results of the tokens with top_p probability mass.
n	Integer	No	1	Number of chat completion choices to generate for each input.
stream	Boolean	No	`false`	Whether to stream responses. This specific endpoint is defined as non-streaming.
max_tokens	Integer	No	inf	The maximum number of tokens to generate.
stop	String/Array	No	null	Stop sequences. The API will stop generating further tokens if these are encountered.
presence_penalty	Number	No	0	Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far.
frequency_penalty	Number	No	0	Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text.
response_format	Object	No	-	Specifies the output format, e.g., `{ "type": "json_object" }` to enable JSON mode.
user	String	No	-	A unique identifier representing your end-user, used to monitor abuse.
seed	Integer	No	-	(Beta) If specified, the system will make a best effort to sample deterministically.

Response Parameters

Parameter Name	Type	Description
id	String	A unique identifier for the request.
object	String	The object type, always `chat.completion`.
created	Integer	The Unix timestamp (in seconds) of when the completion was created.
choices	Array	A list of chat completion choices.
└ index	Integer	The index of the choice.
└ message	Object	A message object containing `role` and `content`.
└ finish_reason	String	The reason the model stopped generating (e.g., `stop`, `length`).
usage	Object	Usage statistics for the completion request.
└ prompt_tokens	Integer	Number of tokens in the prompt.
└ completion_tokens	Integer	Number of tokens in the generated completion.
└ total_tokens	Integer	Total number of tokens used in the request (prompt + completion).

Examples

Request Example (Image Enhancement/Creation)

{
    "model": "gpt-4o-image-vip",
    "messages": [
      {
            "role": "user",
            "content": [
                {"type": "text", "text": "Beautify this image, add the text 'I Love China', and set the aspect ratio to [4:3]"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://lsky.zhongzhuan.chat/i/2024/10/17/6711068a14527.png"
                    }
                }
            ]
        }
    ],
    "tools": [],
    "tool_choice": "none"
}

Response Example

{
    "id": "chatcmpl-123",
    "object": "chat.completion",
    "created": 1677652288,
    "choices": [
        {
            "index": 0,
            "message": {
                "role": "assistant",
                "content": "Sure, I have processed the image for you. You can view the generated result here: [Image Link/Description]"
            },
            "finish_reason": "stop"
        }
    ],
    "usage": {
        "prompt_tokens": 9,
        "completion_tokens": 12,
        "total_tokens": 21
    }
}

FAQs

Q: Why does the request fail when using response_format: { "type": "json_object" }? A: When using JSON mode, you must explicitly instruct the model to generate JSON in your messages (e.g., by including "Please respond in JSON format" in the system prompt). Otherwise, the model may produce non-JSON content, leading to errors or infinite loops.

Q: How can I ensure consistent results across requests? A: While AI models are inherently probabilistic, you can try setting the seed parameter and keeping temperature at 0. Additionally, monitor the system_fingerprint field in the response to track if any backend changes have occurred.

Q: What is the recommended value for max_tokens? A: This depends on your use case. For short image descriptions, 100-300 tokens are usually sufficient. For detailed analysis or creative writing, we recommend 1000 or more. Note that the sum of input tokens (including images) and output tokens cannot exceed the model's context limit.

Q: How do I handle image inputs? A: Use type: "image_url" within the content array of the messages. We currently recommend using publicly accessible HTTPS image links.

Q: Are tools and tool_choice mandatory? A: According to this interface definition, these two fields are marked as required. If you do not need to call specific functions, please set tools to an empty array [] and tool_choice to "none".