cURL

curl --request POST \
  --url https://api.z.ai/api/v1/agents \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "agent_id": "general_translation",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "<string>"
        }
      ]
    }
  ],
  "stream": true,
  "custom_variables": {
    "glossary": "<string>",
    "strategy_config": {
      "general": {
        "suggestion": "<string>"
      },
      "cot": {}
    }
  }
}
'

import requests

url = "https://api.z.ai/api/v1/agents"

payload = {
    "agent_id": "general_translation",
    "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": "<string>"
                }
            ]
        }
    ],
    "stream": True,
    "custom_variables": {
        "glossary": "<string>",
        "strategy_config": {
            "general": { "suggestion": "<string>" },
            "cot": {}
        }
    }
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    agent_id: 'general_translation',
    messages: [{role: 'user', content: [{type: 'text', text: '<string>'}]}],
    stream: true,
    custom_variables: {
      glossary: '<string>',
      strategy_config: {general: {suggestion: '<string>'}, cot: {}}
    }
  })
};

fetch('https://api.z.ai/api/v1/agents', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

HttpResponse<String> response = Unirest.post("https://api.z.ai/api/v1/agents")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"agent_id\": \"general_translation\",\n  \"messages\": [\n    {\n      \"role\": \"user\",\n      \"content\": [\n        {\n          \"type\": \"text\",\n          \"text\": \"<string>\"\n        }\n      ]\n    }\n  ],\n  \"stream\": true,\n  \"custom_variables\": {\n    \"glossary\": \"<string>\",\n    \"strategy_config\": {\n      \"general\": {\n        \"suggestion\": \"<string>\"\n      },\n      \"cot\": {}\n    }\n  }\n}")
  .asString();

{
  "id": "<string>",
  "agent_id": "<string>",
  "choices": [
    {
      "index": 123,
      "finish_reason": "<string>",
      "messages": [
        {
          "role": "<string>",
          "content": {
            "text": "<string>",
            "type": "<string>"
          }
        }
      ]
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123,
    "total_calls": 123
  }
}

Agent API

Agent Chat

General Translation: General Translation API provides large model-based multilingual translation services, including general translation, paraphrase translation, two-step translation, and three-pass translation strategies. It supports automatic language detection, glossary customization, translation suggestions, and streaming output. Users only need to call the Translation API, input the text to be processed, specify the source language (auto-detection supported) and target language to receive high-quality translation results.

Popular Special Effects Videos: Popular special effects videos are intelligent templates launched based on trending features from pan-entertainment platforms, designed to precisely adapt to short video creative production needs. Currently, three effect templates are available: French Kiss, Body Shake Dance, and Sexy Me. After selecting a template, users only need to upload an image and enter corresponding prompts to generate a special effects video.

GLM Slide/Poster Agent: An intelligent creation agent built for working people and creators. It goes beyond traditional engineering-style assembly tools—supporting one-click generation of slides or posters from natural language instructions. By natively integrating content generation with layout aesthetics and design conventions, it helps you quickly produce polished, professional-grade materials while lowering design barriers and boosting creative efficiency.

POST

agents

cURL

curl --request POST \
  --url https://api.z.ai/api/v1/agents \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "agent_id": "general_translation",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "<string>"
        }
      ]
    }
  ],
  "stream": true,
  "custom_variables": {
    "glossary": "<string>",
    "strategy_config": {
      "general": {
        "suggestion": "<string>"
      },
      "cot": {}
    }
  }
}
'

import requests

url = "https://api.z.ai/api/v1/agents"

payload = {
    "agent_id": "general_translation",
    "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": "<string>"
                }
            ]
        }
    ],
    "stream": True,
    "custom_variables": {
        "glossary": "<string>",
        "strategy_config": {
            "general": { "suggestion": "<string>" },
            "cot": {}
        }
    }
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    agent_id: 'general_translation',
    messages: [{role: 'user', content: [{type: 'text', text: '<string>'}]}],
    stream: true,
    custom_variables: {
      glossary: '<string>',
      strategy_config: {general: {suggestion: '<string>'}, cot: {}}
    }
  })
};

fetch('https://api.z.ai/api/v1/agents', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

HttpResponse<String> response = Unirest.post("https://api.z.ai/api/v1/agents")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"agent_id\": \"general_translation\",\n  \"messages\": [\n    {\n      \"role\": \"user\",\n      \"content\": [\n        {\n          \"type\": \"text\",\n          \"text\": \"<string>\"\n        }\n      ]\n    }\n  ],\n  \"stream\": true,\n  \"custom_variables\": {\n    \"glossary\": \"<string>\",\n    \"strategy_config\": {\n      \"general\": {\n        \"suggestion\": \"<string>\"\n      },\n      \"cot\": {}\n    }\n  }\n}")
  .asString();

{
  "id": "<string>",
  "agent_id": "<string>",
  "choices": [
    {
      "index": 123,
      "finish_reason": "<string>",
      "messages": [
        {
          "role": "<string>",
          "content": {
            "text": "<string>",
            "type": "<string>"
          }
        }
      ]
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123,
    "total_calls": 123
  }
}

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

General Translation
Popular Special Effects Videos
GLM Slide/Poster

agent_id

enum<string>

required

Agent ID: general_translation.

Available options:

general_translation

messages

object[]

required

Session message body.

Hide child attributes

messages.role

enum<string>

default:user

required

User input role: user

Available options:

user

Example:

"user"

messages.content

object[]

required

Content list.

Hide child attributes

messages.content.type

enum<string>

default:text

required

Supported type: text.

Available options:

text

messages.content.text

string

required

User text input.

stream

boolean

False for sync calls (default). True for streaming.

custom_variables

object

Hide child attributes

custom_variables.source_lang

enum<string>

Supported language codes (default: auto):

auto: Auto Detect
zh-CN: Simplified Chinese
zh-TW: Traditional Chinese
wyw: Classical Chinese
yue: Cantonese
en: English
ja: Japanese
ko: Korean
fr: French
de: German
es: Spanish
ru: Russian
pt: Portuguese
it: Italian
ar: Arabic
hi: Hindi
bg: Bulgarian
cs: Czech
da: Danish
el: Greek
et: Estonian
fi: Finnish
hu: Hungarian
id: Indonesian
lt: Lithuanian
lv: Latvian
nl: Dutch
no: Norwegian Bokmål
pl: Polish
ro: Romanian
sk: Slovak
sl: Slovenian
sv: Swedish
th: Thai
tr: Turkish
uk: Ukrainian
vi: Vietnamese
my: Burmese
ms: Malay
Pinyin: Pinyin
IPA: International Phonetic Alphabet

Available options:

auto,

zh-CN,

zh-TW,

wyw,

yue,

en,

ja,

ko,

fr,

de,

es,

ru,

pt,

it,

ar,

hi,

bg,

cs,

da,

el,

et,

fi,

hu,

id,

lt,

lv,

nl,

no,

pl,

ro,

sk,

sl,

sv,

th,

tr,

uk,

vi,

my,

ms,

Pinyin,

IPA

custom_variables.target_lang

enum<string>

Target language code (default: zh-CN):

zh-CN: Simplified Chinese
zh-TW: Traditional Chinese
wyw: Classical Chinese
yue: Cantonese
en: English
en-GB: English (British)
en-US: English (American)
ja: Japanese
ko: Korean
fr: French
de: German
es: Spanish
ru: Russian
pt: Portuguese
it: Italian
ar: Arabic
hi: Hindi
bg: Bulgarian
cs: Czech
da: Danish
el: Greek
et: Estonian
fi: Finnish
hu: Hungarian
id: Indonesian
lt: Lithuanian
lv: Latvian
nl: Dutch
no: Norwegian Bokmål
pl: Polish
ro: Romanian
sk: Slovak
sl: Slovenian
sv: Swedish
th: Thai
tr: Turkish
uk: Ukrainian
vi: Vietnamese
my: Burmese
ms: Malay
Pinyin: Pinyin
IPA: International Phonetic Alphabet .

Available options:

zh-CN,

zh-TW,

wyw,

yue,

en,

en-GB,

en-US,

ja,

ko,

fr,

de,

es,

ru,

pt,

it,

ar,

hi,

bg,

cs,

da,

el,

et,

fi,

hu,

id,

lt,

lv,

nl,

no,

pl,

ro,

sk,

sl,

sv,

th,

tr,

uk,

vi,

my,

ms,

Pinyin,

IPA

custom_variables.glossary

string

Glossary ID.

custom_variables.strategy

enum<string>

Translation strategy (default: general)，Optional:

general: General Translation
paraphrase: Paraphrase Translation
two_step: Two-Step Translation
three_step: Three-Stage Translation
reflection: Reflection Translation; cot: COT Translation

Available options:

general,

paraphrase,

two_step,

three_step,

reflection

custom_variables.strategy_config

object

Strategy parameters.

Hide child attributes

custom_variables.strategy_config.general

object

Hide child attributes

custom_variables.strategy_config.general.suggestion

string

Translation suggestions or style requirements (e.g., terminology mapping, style guidelines).

custom_variables.strategy_config.cot

object

Parameters when strategy is cot.

Hide child attributes

custom_variables.strategy_config.cot.reason_lang

enum<string>

Language for translation reasoning, values: [from｜to], default: to.

Available options:

from,

to

Response

Processing successful

General Translation
Popular Special Effects Videos
GLM Slide/Poster

string

Task ID.

agent_id

string

Agent ID.

status

string

Task status.

choices

object[]

Model output content.

Hide child attributes

choices.index

integer

Result index.

choices.finish_reason

string

Termination reason: stop (normal completion), tool_calls (model calls), length (token limit exceeded), sensitive (content flagged), network_error (model inference error).

choices.messages

object

Model response message.

Hide child attributes

choices.messages.role

string

Dialog role (default: assistant).

choices.messages.content

object

Inference result

Hide child attributes

choices.messages.content.type

string

Result type.

choices.messages.content.text

string

Result content.

usage

object

Token usage statistics.

Hide child attributes

usage.prompt_tokens

integer

Input tokens count.

usage.completion_tokens

integer

Output tokens count.

usage.total_tokens

integer

Total tokens count.

usage.total_calls

integer

Total number of calls

Web Reader File Upload