Chat Completion

Create a chat completion model that generates AI replies for given conversation messages. It supports multimodal inputs (text, images, audio, video, file), offers configurable parameters (like temperature, max tokens, tool use), and supports both streaming and non-streaming output modes.

POST /paas/v4/chat/completions

cURL

Basic Example

curl --request POST \
  --url https://api.z.ai/api/paas/v4/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "glm-5.3",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful coding assistant."
    },
    {
      "role": "user",
      "content": "Write a Python function to calculate the factorial of a number."
    }
  ],
  "temperature": 1,
  "stream": false
}
'

200

{
  "id": "<string>",
  "request_id": "<string>",
  "created": 123,
  "model": "<string>",
  "choices": [
    {
      "index": 123,
      "message": {
        "role": "assistant",
        "content": "<string>",
        "reasoning_content": "<string>",
        "tool_calls": [
          {
            "function": {
              "name": "<string>",
              "arguments": {}
            },
            "id": "<string>",
            "type": "<string>"
          }
        ]
      },
      "finish_reason": "<string>"
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "prompt_tokens_details": {
      "cached_tokens": 123
    },
    "total_tokens": 123
  },
  "web_search": [
    {
      "title": "<string>",
      "content": "<string>",
      "link": "<string>",
      "media": "<string>",
      "icon": "<string>",
      "refer": "<string>",
      "publish_date": "<string>"
    }
  ]
}

Authorizations

Authorization · string · header · required

Use the following format for authentication: Bearer

Headers

Accept-Language · enum<string> · default: en-US,en

Config desired response language for HTTP requests.

Available options: en-US,en

Example:

"en-US,en"

Body

application/json

Text Model

Vision Model

model · enum<string> · default: glm-5.3 · required

The model code to be called. GLM-5.3, GLM-5.2, GLM-5.1 are the latest flagship model series, foundational models specifically designed for agent applications.

Available options: glm-5.3, glm-5.2, glm-5.1, glm-5, glm-4.7, glm-4.7-flash, glm-4.7-flashx, glm-4.6, glm-4.5, glm-4.5-air, glm-4.5-x, glm-4.5-airx, glm-4.5-flash, glm-4-32b-0414-128k

Example:

"glm-5.3"

messages · (User Message · object | System Message · object | Assistant Message · object | Tool Message · object)[] · required

The current conversation message list as the model’s prompt input, provided in JSON array format, e.g.,{“role”: “user”, “content”: “Hello”}. Possible message types include system messages, user messages, assistant messages, and tool messages. Note: The input must not consist of system messages or assistant messages only.

Minimum array length: 1

User Message

System Message

Assistant Message

Tool Message

Hide child attributes

messages. role · enum<string> · default: user · required

Role of the message author

Available options: user

messages. content · string · required

Text message content

Example:

"What opportunities and challenges will the Chinese large model industry face in 2025?"

do_sample · boolean · default: true

When do_sample is true, sampling strategy is enabled; when do_sample is false, sampling strategy parameters such as temperature and top_p will not take effect. Default value is true.

Example:

true

stream · boolean · default: false

This parameter should be set to false or omitted when using synchronous call. It indicates that the model returns all content at once after generating all content. Default value is false. If set to true, the model will return the generated content in chunks via standard Event Stream. When the Event Stream ends, a data: [DONE] message will be returned.

Example:

false

thinking · object

Only supported by GLM-4.5 series and higher models. This parameter is used to control whether the model enable the chain of thought.

Hide child attributes

thinking. type · enum<string> · default: enabled

Whether to enable the chain of thought(GLM-5.3 GLM-5.3-FLASH can only be enabled, and the thinking depth is controlled by reasoning_effort; for other models, when enabled, GLM-5.2 GLM-5.1 GLM-5 GLM-4.6 GLM-4.5 and others will automatically determine whether to think, while GLM-4.7 and GLM-4.5V will think compulsorily), default: enabled

Available options: enabled, disabled

thinking. clear_thinking · boolean · default: true

Default value is True. Controls whether to clear reasoning_content from previous conversation turns. View more in Thinking Mode.

• true (default): For this request, the system ignores/removes reasoning_content from prior turns, and only keeps non-reasoning context (e.g., user/assistant visible text, tool calls, and tool results). This is recommended for general chat or lightweight tasks to reduce context length and cost.
• false: Retains reasoning_content from prior turns and includes it in the context sent to the model. To enable Preserved Thinking, you must forward the full, unmodified, and correctly ordered historical reasoning_content in messages. Missing, truncated, rewritten, or reordered blocks may degrade performance or prevent the feature from taking effect.
• Notes: This parameter only affects cross-turn historical thinking blocks; it does not change whether the model generates/returns thinking in the current turn.

Example:

true

reasoning_effort · enum<string> · default: max

Controls the model's reasoning effort level, takes effect when thinking is enabled. Default is max, supported by GLM-5.2 and above. For the GLM-5.3 GLM-5.3-FLASH model, only the low / high / max levels are supported. For the GLM-5.2 model, for compatibility with other protocols, passing none or minimal will cause the model to skip thinking; low and medium will be mapped to high; xhigh will be mapped to max.

Available options: max, xhigh, high, medium, low, minimal, none

Example:

"max"

temperature · number<float> · default: 1

Sampling temperature, controls the randomness of the output, must be a positive number within the range: [0.0, 1.0]. The GLM-5.3, GLM-5.2, GLM-5.1, GLM-5, GLM-4.7, GLM-4.6 series default value is 1.0, GLM-4.5 series default value is 0.6, GLM-4-32B-0414-128K default value is 0.75.

Required range: 0 <= x <= 1

Example:

1

top_p · number<float> · default: 0.95

Another method of temperature sampling, value range is: [0.01, 1.0]. The GLM-5.3, GLM-5.2, GLM-5.1, GLM-5, GLM-4.7, GLM-4.6, GLM-4.5 series default value is 0.95, GLM-4-32B-0414-128K default value is 0.9.

Required range: 0.01 <= x <= 1

Example:

0.95

max_tokens · integer

The maximum number of tokens for model output, the GLM-5.3, GLM-5.2, GLM-5.1, GLM-5, GLM-4.7, GLM-4.6 series supports 128K maximum output, the GLM-4.5 series supports 96K maximum output, the GLM-4.6v series supports 32K maximum output, the GLM-4.5v series supports 16K maximum output, GLM-4-32B-0414-128K supports 16K maximum output.

Required range: 1 <= x <= 131072

Example:

1024

tool_stream · boolean · default: false

Whether to enable streaming response for Function Calls. Default value is false. Only supported by the GLM-5.3, GLM-5.2, GLM-5.1, GLM-5, GLM-4.7, and GLM-4.6 series. Refer to the Stream Tool Call

Example:

false

tools · (Function Call · object | Retrieval · object | Web Search · object)[]

A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported.

Function Call

Retrieval

Web Search

Hide child attributes

tools. type · enum<string> · default: function · required

Available options: function

tools. function · object · required

Hide child attributes

tools.function. name · string · required

The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.

Required string length: 1 - 64

Pattern: ^[a-zA-Z0-9_-]+$

tools.function. description · string · required

A description of what the function does, used by the model to choose when and how to call the function.

tools.function. parameters · object · required

Parameters defined using JSON Schema. Must pass a JSON Schema object to accurately define accepted parameters. Omit if no parameters are needed when calling the function.

tool_choice · enum<string>

Controls how the model selects a tool.

Available options: auto

stop · string[]

Stop word list. Generation stops when the model encounters any specified string. Currently, only one stop word is supported, in the format ["stop_word1"].

Maximum array length: 4

response_format · object

Specifies the response format of the model. Defaults to text. Only text models support this field. type converges to three values: text (plain text output), json_object (JSON output).

Hide child attributes

response_format. type · enum<string> · default: text · required

Output format type: text for plain text, json_object for JSON-formatted output.

Available options: text, json_object

request_id · string

Passed by the user side, needs to be unique; used to distinguish each request, 6–64 characters. If not provided by the user side, the platform will generate one by default.

Required string length: 6 - 64

user_id · string

Unique ID for the end user, 6–128 characters. Avoid using sensitive information.

Required string length: 6 - 128

Response

200

application/json

Processing successful

id · string

Task ID

request_id · string

Request ID

created · integer

Request creation time, Unix timestamp in seconds

model · string

Model name

choices · object[]

List of model responses

Hide child attributes

choices. index · integer

Result index.

choices. message · object

Hide child attributes

choices.message. role · string

Current conversation role, default is ‘assistant’ (model)

Example:

"assistant"

choices.message. content · string

Current conversation content. Hits function is null, otherwise returns model inference result. For the GLM-4.5V series models, the output may contain the reasoning process tags <think> </think> or the text boundary tags <|begin_of_box|> <|end_of_box|>.

choices.message. reasoning_content · string

Reasoning content, supports by GLM-4.5 series.

choices.message. tool_calls · object[]

Function names and parameters generated by the model that should be called.

Hide child attributes

choices.message.tool_calls. function · object

Contains the function name and JSON format parameters generated by the model.

Hide child attributes

choices.message.tool_calls.function. name · string · required

Model-generated function name.

choices.message.tool_calls.function. arguments · object · required

JSON format of the function call parameters generated by the model. Validate the parameters before calling the function.

choices.message.tool_calls. id · string

Unique identifier for the hit function.

choices.message.tool_calls. type · string

Tool type called by the model, currently only supports ‘function’.

choices. finish_reason · string

Reason for model inference termination. Can be stop, tool_calls, length, sensitive, model_context_window_exceeded or network_error.

usage · object

Token usage statistics returned when the model call ends.

Hide child attributes

usage. prompt_tokens · number

Number of tokens in user input

usage. completion_tokens · number

Number of output tokens

usage. prompt_tokens_details · object

Hide child attributes

usage.prompt_tokens_details. cached_tokens · number

Number of tokens served from cache

usage. total_tokens · integer

Total number of tokens

web_search · object[]

Search results.

Hide child attributes

web_search. title · string

Title.

web_search. content · string

Content summary.

web_search. link · string

Result URL.

web_search. media · string

Website name.

web_search. icon · string

Website icon.

web_search. refer · string

Index number.

web_search. publish_date · string

Website publication date.