philipkiely

How to build function calling and JSON mode for open-source and fine-tuned LLMs

1 point by philipkiely 9 months ago | 0 comments

How to double tokens per second for Llama 3 with Medusa

2 points by philipkiely 10 months ago | 0 comments

FP8: Efficient model inference with 8-bit floating point numbers

2 points by philipkiely 1 year ago | 0 comments

Three techniques to adapt LLMs for any use case

1 point by philipkiely 2 years ago | 0 comments

Serving four million Riffusion requests in two days

5 points by philipkiely 2 years ago | 0 comments

Show HN: Free Stable Diffusion 2.0 hosted interface

25 points by philipkiely 2 years ago | 2 comments

Try it yourself: Speech to text with Whisper

5 points by philipkiely 2 years ago | 0 comments

Deploying Stable Diffusion in Production Using Truss

3 points by philipkiely 2 years ago | 0 comments

Hosted Stable Diffusion Demo

7 points by philipkiely 2 years ago | 0 comments

Show HN: Truss – Serve any ML model without boilerplate code

68 points by philipkiely 2 years ago | 9 comments

Code generation interactive demo (Salesforce Codegen mono 2B)

2 points by philipkiely 3 years ago | 0 comments