My model is a 10 billion parameter LLM designed by Acme, Inc...
It is available under the MIT license
Requires an A10 for inference, expected to take 5 seconds per thousand tokens
- Install Truss
- Make an API key
- Deploy to Baseten
Here are some links to docs about stuff:
- Model resources
- Model health
- Model logs
cURL https://baseten.co/endpoint
(Table of parameters)
MyModel.predict("prompt")