LLM Kit Together AI

Together AI provider for LLM Kit - Complete integration with Together AI's extensive collection of open-source models.

Note: This provider uses the standardized builder pattern. See the Quick Start section for the recommended usage.

Features

Text Generation: Generate text using Llama, Mistral, Qwen, DeepSeek, and more
Streaming: Stream responses in real-time with support for tool calls
Tool Calling: Support for function calling with chat models
Text Embedding: Generate embeddings for semantic search and similarity
Image Generation: Create images with FLUX and Stable Diffusion models
Reranking: Improve search results with document reranking models
Multiple Model Families: Access to Llama, Mistral, Qwen, DeepSeek, Gemma, and more

Installation

Add this to your Cargo.toml:

[dependencies]
llm-kit-togetherai = "0.1"
llm-kit-core = "0.1"
llm-kit-provider = "0.1"
tokio = { version = "1", features = ["full"] }

Quick Start

Using the Client Builder (Recommended)

use llm_kit_togetherai::TogetherAIClient;
use llm_kit_provider::language_model::LanguageModel;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create provider using the client builder
    let provider = TogetherAIClient::new()
        .api_key("your-api-key")  // Or use TOGETHER_AI_API_KEY env var
        .build();
    
    // Create a language model
    let model = provider.chat_model("meta-llama/Llama-3.3-70B-Instruct-Turbo");
    
    println!("Model: {}", model.model_id());
    println!("Provider: {}", model.provider());
    Ok(())
}

Using Settings Directly (Alternative)

use llm_kit_togetherai::{TogetherAIProvider, TogetherAIProviderSettings};
use llm_kit_provider::language_model::LanguageModel;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create provider with settings
    let provider = TogetherAIProvider::new(
        TogetherAIProviderSettings::new()
            .with_api_key("your-api-key")
    );
    
    let model = provider.chat_model("meta-llama/Llama-3.3-70B-Instruct-Turbo");
    
    println!("Model: {}", model.model_id());
    Ok(())
}

Configuration

Environment Variables

Set your Together AI API key as an environment variable:

export TOGETHER_AI_API_KEY=your-api-key

Using the Client Builder

use llm_kit_togetherai::TogetherAIClient;

let provider = TogetherAIClient::new()
    .api_key("your-api-key")
    .base_url("https://api.together.xyz/v1")
    .header("Custom-Header", "value")
    .name("my-togetherai")
    .build();

Builder Methods

The TogetherAIClient builder supports:

.api_key(key) - Set the API key (overrides TOGETHER_AI_API_KEY environment variable)
.base_url(url) - Set custom base URL (default: https://api.together.xyz/v1)
.name(name) - Set provider name (optional)
.header(key, value) - Add a single custom header
.headers(map) - Add multiple custom headers
.load_api_key_from_env() - Explicitly load API key from environment variable
.build() - Build the provider

Supported Models

Chat Models

All Together AI chat models are supported, including:

Llama: meta-llama/Llama-3.3-70B-Instruct-Turbo, meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo, meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo, meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
Qwen: Qwen/Qwen2.5-Coder-32B-Instruct, Qwen/Qwen2.5-7B-Instruct-Turbo, Qwen/Qwen2.5-72B-Instruct-Turbo
DeepSeek: deepseek-ai/DeepSeek-V3
Mistral: mistralai/Mistral-7B-Instruct-v0.3, mistralai/Mixtral-8x7B-Instruct-v0.1, mistralai/Mixtral-8x22B-Instruct-v0.1
Gemma: google/gemma-2-9b-it, google/gemma-2-27b-it

Embedding Models

WhereIsAI: WhereIsAI/UAE-Large-V1
BAAI: BAAI/bge-large-en-v1.5, BAAI/bge-base-en-v1.5
Sentence Transformers: sentence-transformers/msmarco-bert-base-dot-v5

Image Models

FLUX: black-forest-labs/FLUX.1-schnell, black-forest-labs/FLUX.1-dev, black-forest-labs/FLUX.1.1-pro
Stable Diffusion: stabilityai/stable-diffusion-xl-base-1.0, stabilityai/stable-diffusion-2-1

Reranking Models

Salesforce: Salesforce/Llama-Rank-v1
Mixedbread: mixedbread-ai/Mxbai-Rerank-Large-V2

For a complete list of available models, see the Together AI Models documentation.

Provider-Specific Options

Chained Model Creation

Together AI provider supports convenient chained model creation:

use llm_kit_togetherai::TogetherAIClient;

// Create model directly from builder
let model = TogetherAIClient::new()
    .api_key("your-api-key")
    .build()
    .chat_model("meta-llama/Llama-3.3-70B-Instruct-Turbo");

Multiple Model Types

Together AI supports multiple model types in a single provider:

use llm_kit_togetherai::TogetherAIClient;

let provider = TogetherAIClient::new()
    .api_key("your-api-key")
    .build();

// Chat models
let chat_model = provider.chat_model("meta-llama/Llama-3.3-70B-Instruct-Turbo");

// Embedding models
let embedding_model = provider.text_embedding_model("WhereIsAI/UAE-Large-V1");

// Image models
let image_model = provider.image_model("black-forest-labs/FLUX.1-schnell");

// Reranking models
let reranking_model = provider.reranking_model("Salesforce/Llama-Rank-v1");

Examples

See the examples/ directory for complete examples:

chat.rs - Basic chat completion
stream.rs - Streaming responses
chat_tool_calling.rs - Tool calling with function definitions
stream_tool_calling.rs - Streaming with tool calls
text_embedding.rs - Text embeddings for semantic search
image_generation.rs - Image generation with FLUX and Stable Diffusion
reranking.rs - Document reranking for improved search

Run examples with:

export TOGETHER_AI_API_KEY=your-api-key
cargo run --example chat
cargo run --example stream
cargo run --example chat_tool_calling
cargo run --example stream_tool_calling
cargo run --example text_embedding
cargo run --example image_generation
cargo run --example reranking

llm-kit-togetherai 0.1.0