Zhipu BigModel

https://docs.bigmodel.cn/cn/
Admin
Zhipu BigModel is a one-stop large language model development platform providing multimodal AI APIs,...
Tokens:4,353
Snippets:16
Trust Score:8.8
Update:2 weeks ago
Show doc for...
Tokens:
Raw
### Verify zai-sdk installation

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Verify the zai-sdk installation by importing the package and printing its version.

```python
import zai
print(zai.__version__)
```

--------------------------------

### Install zai-sdk Python package

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Install the zai-sdk Python package, either the latest version or a specific version (0.2.3).

```bash
# 安装最新版本
pip install zai-sdk
# 或指定版本
pip install zai-sdk==0.2.3
```

--------------------------------

### zai-sdk Java: Installation (Maven)

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Maven dependency for the zai-sdk Java library.

```APIDOC
## zai-sdk Java: Installation (Maven)

### Description
Maven dependency for the zai-sdk Java library.

### Dependency
```xml
<dependency>
    <groupId>ai.z.openapi</groupId>
    <artifactId>zai-sdk</artifactId>
    <version>0.3.5</version>
</dependency>
```
```

--------------------------------

### zai-sdk Java: Installation (Gradle)

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Gradle dependency for the zai-sdk Java library.

```APIDOC
## zai-sdk Java: Installation (Gradle)

### Description
Gradle dependency for the zai-sdk Java library.

### Dependency
```groovy
implementation 'ai.z.openapi:zai-sdk:0.3.5'
```
```

--------------------------------

### zai-sdk Python: chat.completions.create (Basic)

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Python SDK method to create a chat completion with the GLM-4.6V-Flash model, enabling thinking mode.

```APIDOC
## zai-sdk Python: chat.completions.create (Basic)

### Description
Python SDK method to create a chat completion with the GLM-4.6V-Flash model, enabling thinking mode.

### Method
create

### Endpoint
client.chat.completions.create

### Parameters
- **model** (string) - Required - The model identifier, e.g., "glm-4.6v-flash".
- **messages** (array) - Required - A list of message objects. Each message has a role and content.
  - **role** (string) - Required - The role of the message, e.g., "user".
  - **content** (array) - Required - A list of content parts. Each part has a type and associated data.
    - **type** (string) - Required - The type of content, e.g., "image_url" or "text".
    - **image_url** (object) - Required when type is "image_url" - Contains the image URL.
      - **url** (string) - Required - The URL of the image.
    - **text** (string) - Required when type is "text" - The text prompt.
- **thinking** (dict) - Optional - Configuration for thinking mode.
  - **type** (string) - Required - The thinking mode type, e.g., "enabled".

### Request Example
```python
from zai import ZhipuAiClient

client = ZhipuAiClient(api_key="YOUR_API_KEY")
response = client.chat.completions.create(
    model="glm-4.6v-flash",
    messages=[
        {
            "content": [
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
                    }
                },
                {
                    "type": "text",
                    "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
                }
            ],
            "role": "user"
        }
    ],
    thinking={
        "type": "enabled"
    }
)
print(response.choices[0].message)
```

### Response
#### Success Response
- **choices** (list) - List of completion choices.
  - **message** (object) - The generated message.

#### Response Example
```python
response.choices[0].message
```
```

--------------------------------

### zai-sdk Python: chat.completions.create (Streaming)

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Python SDK method to create a streaming chat completion with the GLM-4.6V-Flash model, enabling thinking mode and streaming responses.

```APIDOC
## zai-sdk Python: chat.completions.create (Streaming)

### Description
Python SDK method to create a streaming chat completion with the GLM-4.6V-Flash model, enabling thinking mode and streaming responses.

### Method
create

### Endpoint
client.chat.completions.create

### Parameters
- **model** (string) - Required - The model identifier, e.g., "glm-4.6v-flash".
- **messages** (array) - Required - A list of message objects. Each message has a role and content.
  - **role** (string) - Required - The role of the message, e.g., "user".
  - **content** (array) - Required - A list of content parts. Each part has a type and associated data.
    - **type** (string) - Required - The type of content, e.g., "image_url" or "text".
    - **image_url** (object) - Required when type is "image_url" - Contains the image URL.
      - **url** (string) - Required - The URL of the image.
    - **text** (string) - Required when type is "text" - The text prompt.
- **thinking** (dict) - Optional - Configuration for thinking mode.
  - **type** (string) - Required - The thinking mode type, e.g., "enabled".
- **stream** (bool) - Optional - Set to True to enable streaming responses.

### Request Example
```python
from zai import ZhipuAiClient

client = ZhipuAiClient(api_key="YOUR_API_KEY")
response = client.chat.completions.create(
    model="glm-4.6v-flash",
    messages=[
        {
            "content": [
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
                    }
                },
                {
                    "type": "text",
                    "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
                }
            ],
            "role": "user"
        }
    ],
    thinking={
        "type": "enabled"
    },
    stream=True
)

for chunk in response:
    if chunk.choices[0].delta.reasoning_content:
        print(chunk.choices[0].delta.reasoning_content, end='', flush=True)

    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end='', flush=True)
```

### Response
#### Success Response
- **choices** (list) - List of completion choices.
  - **delta** (object) - The streaming delta.
    - **reasoning_content** (string) - The reasoning content, if available.
    - **content** (string) - The generated content, if available.

#### Response Example
```python
for chunk in response:
    if chunk.choices[0].delta.reasoning_content:
        print(chunk.choices[0].delta.reasoning_content, end='', flush=True)
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end='', flush=True)
```
```

--------------------------------

### POST /api/paas/v4/chat/completions (Basic)

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Sends a chat completion request to the GLM-4.6V-Flash model with an image and text prompt, enabling thinking mode.

```APIDOC
## POST /api/paas/v4/chat/completions

### Description
Sends a chat completion request to the GLM-4.6V-Flash model with an image and text prompt, enabling thinking mode.

### Method
POST

### Endpoint
https://open.bigmodel.cn/api/paas/v4/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model identifier, e.g., "glm-4.6v-flash".
- **messages** (array) - Required - A list of message objects. Each message has a role and content.
  - **role** (string) - Required - The role of the message, e.g., "user".
  - **content** (array) - Required - A list of content parts. Each part has a type and associated data.
    - **type** (string) - Required - The type of content, e.g., "image_url" or "text".
    - **image_url** (object) - Required when type is "image_url" - Contains the image URL.
      - **url** (string) - Required - The URL of the image.
    - **text** (string) - Required when type is "text" - The text prompt.
- **thinking** (object) - Optional - Configuration for thinking mode.
  - **type** (string) - Required - The thinking mode type, e.g., "enabled".

### Request Example
```json
{
  "model": "glm-4.6v-flash",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "image_url",
          "image_url": {
            "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
          }
        },
        {
          "type": "text",
          "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
        }
      ]
    }
  ],
  "thinking": {
    "type": "enabled"
  }
}
```

### Response
#### Success Response (200)
- **choices** (array) - List of completion choices.
  - **message** (object) - The generated message.

#### Response Example
```json
{
  "choices": [
    {
      "message": {
        "content": "..."
      }
    }
  ]
}
```
```

--------------------------------

### Gradle dependency for zai-sdk

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Gradle (Groovy) dependency for the zai-sdk Java library, version 0.3.5.

```groovy
implementation 'ai.z.openapi:zai-sdk:0.3.5'
```

--------------------------------

### Python basic call to glm-4.6v-flash

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Basic Python call using ZhipuAiClient to the GLM-4.6V-Flash model with an image and text prompt, enabling thinking. Prints the first choice message.

```python
from zai import ZhipuAiClient

client = ZhipuAiClient(api_key="YOUR_API_KEY")  # 填写您自己的 APIKey
response = client.chat.completions.create(
    model="glm-4.6v-flash",  # 填写需要调用的模型名称
    messages=[
        {
            "content": [
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
                    }
                },
                {
                    "type": "text",
                    "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
                }
            ],
            "role": "user"
        }
    ],
    thinking={
        "type": "enabled"
    }
)
print(response.choices[0].message)
```

--------------------------------

### Maven dependency for zai-sdk

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Maven dependency for the zai-sdk Java library, version 0.3.5.

```xml
<dependency>
    <groupId>ai.z.openapi</groupId>
    <artifactId>zai-sdk</artifactId>
    <version>0.3.5</version>
</dependency>
```

--------------------------------

### POST /api/paas/v4/chat/completions (Streaming)

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Sends a streaming chat completion request to the GLM-4.6V-Flash model, enabling thinking mode and streaming responses.

```APIDOC
## POST /api/paas/v4/chat/completions

### Description
Sends a streaming chat completion request to the GLM-4.6V-Flash model, enabling thinking mode and streaming responses.

### Method
POST

### Endpoint
https://open.bigmodel.cn/api/paas/v4/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model identifier, e.g., "glm-4.6v-flash".
- **messages** (array) - Required - A list of message objects. Each message has a role and content.
  - **role** (string) - Required - The role of the message, e.g., "user".
  - **content** (array) - Required - A list of content parts. Each part has a type and associated data.
    - **type** (string) - Required - The type of content, e.g., "image_url" or "text".
    - **image_url** (object) - Required when type is "image_url" - Contains the image URL.
      - **url** (string) - Required - The URL of the image.
    - **text** (string) - Required when type is "text" - The text prompt.
- **thinking** (object) - Optional - Configuration for thinking mode.
  - **type** (string) - Required - The thinking mode type, e.g., "enabled".
- **stream** (boolean) - Optional - Set to true to enable streaming responses.

### Request Example
```json
{
  "model": "glm-4.6v-flash",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "image_url",
          "image_url": {
            "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
          }
        },
        {
          "type": "text",
          "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
        }
      ]
    }
  ],
  "thinking": {
    "type": "enabled"
  },
  "stream": true
}
```

### Response
#### Success Response (200)
- **choices** (array) - List of completion choices.
  - **delta** (object) - The streaming delta.
    - **reasoning_content** (string) - The reasoning content, if available.
    - **content** (string) - The generated content, if available.

#### Response Example
```json
{
  "choices": [
    {
      "delta": {
        "reasoning_content": "...",
        "content": "..."
      }
    }
  ]
}
```
```

--------------------------------

### cURL streaming call to glm-4.6v-flash

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

cURL request with stream set to true for streaming responses from the GLM-4.6V-Flash model. Same image and prompt as the basic call, but enables incremental output.

```bash
curl -X POST \
https://open.bigmodel.cn/api/paas/v4/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "glm-4.6v-flash",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "image_url",
          "image_url": {
            "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
          }
        },
        {
          "type": "text",
          "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
        }
      ]
    }
  ],
  "thinking": {
    "type": "enabled"
  },
  "stream": true
}'
```

--------------------------------

### Python streaming call to glm-4.6v-flash

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Python streaming call to GLM-4.6V-Flash with stream=True. Iterates over response chunks, printing reasoning_content and content as they arrive.

```python
from zai import ZhipuAiClient

client = ZhipuAiClient(api_key="YOUR_API_KEY")  # 填写您自己的APIKey
response = client.chat.completions.create(
    model="glm-4.6v-flash",  # 填写需要调用的模型名称
    messages=[
        {
            "content": [
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
                    }
                },
                {
                    "type": "text",
                    "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
                }
            ],
            "role": "user"
        }
    ],
    thinking={
        "type": "enabled"
    },
    stream=True
)

for chunk in response:
    if chunk.choices[0].delta.reasoning_content:
        print(chunk.choices[0].delta.reasoning_content, end='', flush=True)

    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end='', flush=True)
```

--------------------------------

### cURL basic call to glm-4.6v-flash

Source: https://docs.bigmodel.cn/cn/guide/models/free/glm-4.6v-flash

Basic cURL request to the GLM-4.6V-Flash chat completions endpoint with an image URL and a text prompt asking for coordinates. Uses the thinking parameter enabled.

```bash
curl -X POST \
https://open.bigmodel.cn/api/paas/v4/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "glm-4.6v-flash",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "image_url",
          "image_url": {
            "url": "https://cloudcovert-1305175928.cos.ap-guangzhou.myqcloud.com/%E5%9B%BE%E7%89%87grounding.PNG"
          }
        },
        {
          "type": "text",
          "text": "Where is the second bottle of beer from the right on the table?  Provide coordinates in [[xmin,ymin,xmax,ymax]] format"
        }
      ]
    }
  ],
  "thinking": {
    "type": "enabled"
  }
}'
```
Tokens:
Raw