Skip to content

Latest commit

 

History

History
14 lines (9 loc) · 784 Bytes

File metadata and controls

14 lines (9 loc) · 784 Bytes

Accuracy Gains Without Training Your Own Model: Parallel Reasoning for LLM Apps

Hack The North 2025

By Eugene Cho

This repo serves as a complement to the workshop slides (see slides.pdf), containing demo code (warning it was vibe coded! use at your own risk!).

Setup

  1. Create a Python virtual environment with version >=3.10. See https://docs.python.org/3/library/venv.html for a tutorial on how to do this. a. Or optionally, if you have the uv Python package manager set up on your device (good for you if you do), you can just run: uv venv, source .venv/bin/activate and then uv sync. :')

  2. Take your Cohere API key, and run export CO_API_KEY=<insert your api key> in the terminal to set your environment variable.

  3. Follow along in toy_problem.ipynb!