Crate novel_zhipu 

Rust library for Zhipu AI (GLM) APIs.

Creating a client

use novel_zhipu::Client;
use novel_zhipu::config::ZhipuConfig;

// Create a client with default configuration from environment variables.
let client = Client::new();

// Or create with custom configuration.
let config = ZhipuConfig::new().with_api_key("your-api-key");
let client = Client::with_config(config);

Chat Completions

use novel_zhipu::Client;
use novel_zhipu::spec::chat::{ChatMessage, CreateChatCompletionRequestArgs, Model};

let client = Client::new();

let request = CreateChatCompletionRequestArgs::default()
    .model(Model::Glm4)
    .messages(vec![
        ChatMessage::system("You are a helpful assistant."),
        ChatMessage::user("Hello!"),
    ])
    .build()?;

let response = client.chat().create(request).await?;
println!("{}", response.choices[0].message.content);

Streaming Chat Completions

use futures_util::StreamExt;
use novel_zhipu::Client;
use novel_zhipu::spec::chat::{ChatMessage, CreateChatCompletionRequestArgs};

let client = Client::new();

let request = CreateChatCompletionRequestArgs::default()
    .model("glm-4")
    .messages(vec![ChatMessage::user("Tell me a story.")])
    .build()?;

let mut stream = client.chat().create_stream(request).await?;
while let Some(chunk) = stream.next().await {
    if let Ok(chunk) = chunk {
        if let Some(content) = &chunk.choices[0].delta.content {
            print!("{}", content);
        }
    }
}

Embeddings

use novel_zhipu::Client;

let client = Client::new();

let embedding = client
    .embeddings()
    .embed_text("embedding-2", "Hello, world!")
    .await?;

println!("Embedding dimension: {}", embedding.len());

Image Generation

use novel_zhipu::Client;

let client = Client::new();

let image = client
    .images()
    .generate("cogview-3", "A beautiful sunset over the ocean")
    .await?;

image.save("sunset.png").await?;

Environment Variables

• ZHIPUAI_API_KEY or ZHIPU_API_KEY: API key
• ZHIPUAI_BASE_URL or ZHIPU_API_BASE: API base URL (default: https://open.bigmodel.cn/api/paas/v4)

Text-to-Speech

use novel_zhipu::Client;

let client = Client::new();

// Simple synthesis
let audio = client.tts().synthesize("你好，今天天气怎么样？").await?;

audio.save("output.wav").await?;

Speech-to-Text

use novel_zhipu::Client;

let client = Client::new();

let result = client.asr().transcribe_file("audio.wav").await?;

println!("Transcription: {}", result.text);

Environment Variables

• ZHIPUAI_API_KEY or ZHIPU_API_KEY: API key
• ZHIPUAI_BASE_URL or ZHIPU_API_BASE: API base URL (default: https://open.bigmodel.cn/api/paas/v4)

Features

• chat: Enable Chat Completions API
• embeddings: Enable Embeddings API
• images: Enable Images API
• tts: Enable Text-to-Speech API
• asr: Enable Speech-to-Text API
• async-task: Enable Async Task APIs (async chat, video, image generation)
• voice: Enable Voice APIs (clone, list, delete)
• reranking: Enable Text Reranking API
• tokenizer: Enable Text Tokenizer API
• tools: Enable Tool APIs (web search, web reader, moderation, file parser)
• full: Enable all features
• rustls: Use rustls for TLS (default)
• native-tls: Use native-tls for TLS

Modules

config

Configuration for Zhipu AI API client.

error

Error types for Zhipu AI API client.

spec

Type definitions for Zhipu AI APIs.

Structs

Agentsagents

Agents API.

Asrasr

Speech-to-text (ASR) API.

Assistantassistant

Assistant API.

AsyncTaskasync-task

Async task API.

Batchesbatch

Batch API.

Chatchat

Chat completions API.

Client

Zhipu AI API client.

Embeddingsembeddings

Embeddings API.

FileParsertools

File parsing API.

Filesfiles

Files API.

Imagesimages

Images API.

Moderationtools

Content moderation API.

Ocrocr

OCR API.

Rerankingreranking

Text reranking API.

Tokenizertokenizer

Text tokenizer API.

Ttstts

Text-to-speech API.

Videosvideos

Videos API.

Voicevoice

Voice API.

WebReadertools

Web reader API.

WebSearchtools

Web search API.