Getting Started
Step-by-step guide to set up and use the Gintonic inference subchain. Learn how to create an account, obtain API keys, and make your first API call.
Last updated
Step-by-step guide to set up and use the Gintonic inference subchain. Learn how to create an account, obtain API keys, and make your first API call.
Last updated
Ready to harness the power of distributed AI inference? Let's get you set up and running in no time.
Now that you've set up your environment, you might be wondering what you can build. For inspiration, check out , a demo application built using the Gintonic inference subchain.
AI Speakeasy features an AI bartender powered by the Mistral model, demonstrating how you can create interactive, chat-based AI applications. It's a great example of what's possible with the technology you've just set up.
As you continue your journey with Gintonic, keep AI Speakeasy in mind as a benchmark for the kind of engaging, responsive AI applications you can create.
Before you start firing off API requests, make sure you've got these basics covered:
Gintonic Inference Account: If you haven't already, sign up at . It's quick, painless, and free to create an account.
API Key: You'll need this to authenticate your requests. Here's how to get one:
Log into your Gintonic account.
Navigate to the 'API Keys' section.
Click 'Generate New Key'.
Copy and save your key somewhere safe. We'll only show it to you once!
Wallet Setup: You'll need a blockchain wallet to handle transactions on the platform.
We recommend using MetaMask. If you don't have it, grab it from
Once installed, connect your wallet to the Gintonic platform.
$GIN: Make sure you have some $GIN in your wallet. These are used to pay for inference requests.
Before you begin, ensure your system meets the following minimum requirements:
Operating System: Linux Ubuntu 20.04 or later.
GPU: Minimum 2G VRAM.
Drivers: Video card drivers and the 'nvidia-smi' package must be installed.
HuggingFace: An account on HuggingFace and an Access Token.
Blockchain: A valid blockchain wallet address.
To set up the inference subchain on your system:
Run the installer. It will automatically:
Install additional dependencies not specified in the prerequisites.
Create a worker that connects to the required swarm.
Calculate the number of model blocks occupied by the worker.
Send the worker's address and ID to the blockchain for future awarding.
Let's make your first API call to the inference subchain:
Install the gintonic SDK
We've got SDKs for popular languages. Let's use Python for this example:
Initialize the client
Start a chat session
Send a message
End the chat session
Congratulations! You've just made your first inference request using the Gintonic inference subchain.
Dive into our API Reference to explore all available endpoints.
Check out Best Practices to optimize your usage.
Got questions? Head over to our FAQs or Troubleshooting sections.
Remember, we're all about making AI inference accessible and affordable. If you run into any roadblocks, don't hesitate to reach out to our support team.
Happy inferencing!
Download the installer from