AI/ML API

Use AI/ML API with access to 300+ models in Kern agents.

AI/ML API is a platform providing unified access to 300+ AI models including Deepseek, Gemini, ChatGPT, and more — with production-grade uptime and high rate limits.

Authentication

Set your AIMLAPI_API_KEY environment variable. Get your key at aimlapi.com.

1export AIMLAPI_API_KEY=***

1setx AIMLAPI_API_KEY ***

Example

Use AI/ML API with your Agent:

1from kern.agent import Agent
2from kern.models.aimlapi import AIMLAPI
3
4agent = Agent(
5    model=AIMLAPI(id="meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo"),
6    markdown=True,
7)
8
9agent.print_response("Explain how black holes are formed.")

Parameters

Parameter	Type	Default	Description
`id`	`str`	`"gpt-4o-mini"`	The id of the model to use
`name`	`str`	`"AIMLAPI"`	The name of the model
`provider`	`str`	`"AIMLAPI"`	The provider of the model
`api_key`	`Optional[str]`	`None`	The API key for AI/ML API (defaults to AIMLAPI_API_KEY env var)
`base_url`	`str`	`"https://api.aimlapi.com/v1"`	The base URL for the AI/ML API
`max_tokens`	`int`	`4096`	Maximum number of tokens to generate

AIMLAPI extends the OpenAI-compatible interface and supports most parameters from the OpenAI model, where applicable.

Available Models

AI/ML API provides access to 300+ models. Some popular models include:

OpenAI Models: gpt-4o, gpt-4o-mini, o1, o1-mini
Anthropic Models: claude-3-5-sonnet-20241022, claude-3-5-haiku-20241022
Meta Models: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo, meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
Google Models: gemini-1.5-pro, gemini-1.5-flash
DeepSeek Models: deepseek-chat, deepseek-reasoner

For a complete list of available models, visit the AI/ML API documentation.