TokenLab MCP server

This is a helper that connects your AI assistant to TokenLab, an online service that runs many different AI models in one place. Once it is connected, your AI can look up which models exist, compare their prices, and send jobs to them for you. It is handy if you use several AI models and want your assistant to pick and run them without you opening lots of websites.

What this MCP server does

You ask your AI something, like which model is cheapest for writing, or ask it to make an image. Your AI uses this helper to talk to TokenLab. The helper sends the request to TokenLab's service and waits for the answer. Then it hands the result back to your AI, which shows it to you in the chat. For bigger jobs like video or music, it gives you a task number you can check on later.

Click to zoom

What you can do with it

• Compare prices across AI models before you pick one
• See which models TokenLab offers and what each one can do
• Run text chats and questions through many different models
• Generate or edit images, video, music, speech and 3D
• Turn audio into text and text into speech
• Create embeddings, rerank results and translate text
• Check on long-running jobs and fetch the finished files

Try asking your AI

• “Which TokenLab models are cheapest for summarising long documents?”
• “Make an image of a red bicycle in a park and save it for me”
• “Turn this meeting recording into text”
• “Translate this paragraph into French using a low-cost model”

What it gives back to you

You get answers in the chat: lists of models with their prices, written replies from a model, or summaries of what happened. Small images and audio come back right in the chat. Bigger files, like video or 3D results, are saved to a folder and you get a path plus the file type and size. Long jobs give you a task number you can ask about again later.

Before you start

Who it's for

People who use several AI models and want their assistant to compare prices and run jobs for them without juggling many websites.