Skip to main content
Language models generate text responses, decide when to call tools, and process conversation context. They handle all text generation and reasoning for your agent. Models vary in speed (response latency), capability (reasoning quality and accuracy), context length (how much conversation history they can process), and features (vision support, function calling, structured output). Hypertic supports 10+ language model providers, allowing you to switch between OpenAI, Anthropic, Google, xAI, and others seamlessly.

Supported Providers

Click on a provider card to see setup and usage examples:

Anthropic

View guide >

Cohere

View guide >

DeepSeek

View guide >

FireworksAI

View guide >

Google Gemini

View guide >

Groq

View guide >

Mistral

View guide >

Moonshot

View guide >

OpenAI

View guide >

OpenRouter

View guide >

Qwen

View guide >

xAI

View guide >

Basic Usage

Here are examples for different providers:

Parameters

Configure your model with these common parameters:
string
required
The model ID used to generate responses, like “gpt-4” or “claude-sonnet-4-5-20250929”.
number
Sampling temperature. Higher values make output more random, lower values make it more deterministic.
number
Upper bound for the number of tokens that can be generated for a response, including visible output tokens.
string
Your API key for authenticating with the provider. Usually set via environment variables.
Example configuration:
Each provider may have additional parameters. Check your provider’s documentation for details.