GLM-4V系列

GLM-4V系列包含3款模型，分别适用于不同的应用场景。

• GLM-4V-Plus:具备卓越的多模态理解能力，可同时处理最多5张图像，并支持视频内容理解，适用于复杂的多媒体分析场景。
• GLM-4V:专注于图像内容理解，能够同时解析最多5张图像，适合于需要深度图像理解的场合。
• GLM-4V-Flash（免费）: 专注于高效的单一图像理解，适用于快速图像解析的场景，例如实时图像分析或批量图像处理。

同步调用

接口请求

请求参数

Messages 格式

模型可接受的消息类型包括 User message、Assistant message ，不同的消息类型格式有所差异。具体如下：

User message

Assistant message

响应参数

请求示例

上传视频 URL

#视频理解示例、上传视频URL
from zhipuai import ZhipuAI
client = ZhipuAI(api_key="YOUR API KEY") # 填写您自己的APIKey
response = client.chat.completions.create(
model="glm-4v-plus",  # 填写需要调用的模型名称
messages=[
{
"role": "user",
"content": [
{
"type": "video_url",
"video_url": {
"url" : "https://sfile.chatglm.cn/testpath/video/xxxxx.mp4"
}
},
{
"type": "text",
"text": "请仔细描述这个视频"
}
]
}
]
)
print(response.choices[0].message)

上传视频 Base64

import base64
from zhipuai import ZhipuAI
video_path = "/Users/YourCompluter/xxxx.mp4"
with open(video_path, 'rb') as video_file:
video_base = base64.b64encode(video_file.read()).decode('utf-8')
client = ZhipuAI(api_key="YOUR API KEY") # 填写您自己的APIKey
response = client.chat.completions.create(
model="glm-4v-plus",  # 填写需要调用的模型名称
messages=[
{
"role": "user",
"content": [
{
"type": "video_url",
"video_url": {
"url" : video_base
}
},
{
"type": "text",
"text": "请仔细描述这个视频"
}
]
}
]
)
print(response.choices[0].message)

上传图片 URL

from zhipuai import ZhipuAI
client = ZhipuAI(api_key="") # 填写您自己的APIKey
response = client.chat.completions.create(
model="glm-4v",  # 填写需要调用的模型名称
messages=[
{
"role": "user",
"content": [
{
"type": "text",
"text": "图里有什么"
},
{
"type": "image_url",
"image_url": {
"url" : "https://img1.baidu.com/it/u=1369931113,3388870256&fm=253&app=138&size=w931&n=0&f=JPEG&fmt=auto?sec=1703696400&t=f3028c7a1dca43a080aeb8239f09cc2f"
}
}
]
}
]
)
print(response.choices[0].message)

上传图片 Base64

import base64
from zhipuai import ZhipuAI
img_path = "/Users/YourCompluter/xxxx.jpeg"
with open(img_path, 'rb') as img_file:
img_base = base64.b64encode(img_file.read()).decode('utf-8')
client = ZhipuAI(api_key="YOUR API KEY") # 填写您自己的APIKey
response = client.chat.completions.create(
model="glm-4v-plus",  # 填写需要调用的模型名称
messages=[
{
"role": "user",
"content": [
{
"type": "image_url",
"image_url": {
"url": img_base
}
},
{
"type": "text",
"text": "请描述这个图片"
}
]
}
]
)
print(response.choices[0].message)

多轮图片对话

from zhipuai import ZhipuAI
client = ZhipuAI(api_key="YOUR API KEY") # 填写您自己的APIKey
response = client.chat.completions.create(
model="glm-4v-plus",  # 填写需要调用的模型名称
messages=[
{
"content": [
{
"image_url": {
"url": "https://cdn.bigmodel.cn/xxxxxx.png"
},
"type": "image_url"
},
{
"text": "图中有什么",
"type": "text"
}
],
"role": "user"
},
{
"content": [
{
"text": "这是一幅描绘自然风景的画，展现了一片宁静的湖泊，湖水呈现出深浅不一的蓝绿色调。湖边长着一棵盛开的樱花树，粉白色的花朵在阳光下显得格外鲜艳。树下有一位身穿蓝色衣服的人坐在船上，似乎正在享受这美好的时光。\n\n背景是连绵起伏的山脉，山体被绿色的植被覆盖，山顶则被云雾缭绕，增添了几分神秘感。远处还可以看到一座小亭子，它坐落在湖边的岩石上，与周围的自然环境和谐相融。\n\n整个画面给人一种宁静、和谐的感觉，仿佛置身于一个世外桃源之中。",
"type": "text"
}
],
"role": "assistant"
},
{
"content": [
{
"image_url": {
"url": "https://cdn.bigmodel.cn/xxxxxxx.jpeg"
},
"type": "image_url"
},
{
"text": "这个图与上面图有什么不一样",
"type": "text"
}
],
"role": "user"
},
{
"content": [
{
"text": "第一张图和第二张图在画面内容、构图、色彩等方面都有很大的不同，形成鲜明的对比。\n\n1.第一张图是风景画，主要描绘了自然风景，如樱花树、湖泊、山脉等，色彩丰富，充满春天的气息，而第二张图是描绘长城在冬季的风景照片，长城在白雪覆盖的山脉中蜿蜒盘旋，整体色调偏冷，给人一种寒冷、宁静的感觉。\n\n2.构图上，第一张图采用对称式构图，以樱花树为中心，两侧的景物相互呼应，形成平衡和谐的画面效果，第二张图则采用对角线构图，长城作为视觉引导线，引导观众的视线穿越整个画面，增加画面的深度和动态感。\n\n整体来看，这两张图在画面内容、色彩、构图和氛围上都呈现出明显的对比，一张是充满生机和色彩的春日风景画，另一张是宁静而神秘的冬日长城风景照。",
"type": "text"
}
],
"role": "assistant"
},
{
"content": [
{
"image_url": {
"url": "https://cdn.bigmodel.cn/xxxxxxx.jpeg"
},
"type": "image_url"
},
{
"text": "这个图与上一张图有什么区别",
"type": "text"
}
],
"role": "user"
}
]
)
print(response.choices[0].message)

响应示例

{
"created": 1703487403,
"id": "8239375684858666781",
"model": "glm-4v-plus",
"request_id": "8239375684858666781",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"content": "图中有一片蓝色的海和蓝天，天空中有白色的云朵。图片的右下角有一个小岛或者岩石，上面长着深绿色的树木。",
"role": "assistant"
}
}
],
"usage": {
"completion_tokens": 37,
"prompt_tokens": 1037,
"total_tokens": 1074
}
}

salida de streaming

响应参数

请求示例

from zhipuai import ZhipuAI
client = ZhipuAI(api_key="") # 请填写您自己的APIKey
response = client.chat.completions.create(
model="glm-4v-plus",  # 填写需要调用的模型名称
messages=[
{
"role": "user", 
"content": [
{
"type": "image_url",
"image_url": {
"url" : "sfile.chatglm.cn/testpath/xxxx.jpg"
}
},
{
"type": "text",
"text": "图里有什么"
}
]
},
],
stream=True,
)
for chunk in response:
print(chunk.choices[0].delta)

响应示例

data: {"id":"8305986882425703351","created":1705476637,"model":"glm-4v-plus","choices":[{"index":0,"delta":{"role":"assistant","content":"下"}}]}
data: {"id":"8305986882425703351","created":1705476637,"model":"glm-4v-plus","choices":[{"index":0,"delta":{"role":"assistant","content":"角"}}]}
data: {"id":"8305986882425703351","created":1705476637,"model":"glm-4v-plus","choices":[{"index":0,"delta":{"role":"assistant","content":"有一个"}}]}
... ...
data: {"id":"8305986882425703351","created":1705476637,"model":"glm-4v-plus","choices":[{"index":0,"delta":{"role":"assistant","content":"树木"}}]}
data: {"id":"8305986882425703351","created":1705476637,"model":"glm-4v-plus","choices":[{"index":0,"delta":{"role":"assistant","content":"。"}}]}
data: {"id":"8305986882425703351","created":1705476637,"model":"glm-4v-plus","choices":[{"index":0,"finish_reason":"stop","delta":{"role":"assistant","content":""}}],"usage":{"prompt_tokens":1037,"completion_tokens":37,"total_tokens":1074}}