Curated list of diverse type of content related to Chainforge's ecosystem.
Chainforge is an open-source visual programming environment for battle-testing prompts to LLMs, where you can evaluate the robustness of prompts and foundation models in a way that goes beyond anecdotal evidence.
Official ressources serving as the basis to the project
- Live Demo - ⛓️ Off-the-shelf live demo.
- Homepage - 🏠 Main website.
- Documentation - 📖 Documentation website.
- Repository -
The GitHub repository that started it all! - Arxiv Paper - 🧑🎓 HCI (Human Computer Interface) research paper presenting in-lab and interview studies on people testing this tool.
YouTube videos talking about Chainforge (newest first)
-
ChainForge: Demo and Tool Walkthrough - (Jan 2025, 35min, by its creator) High-quality and in-depth presentation.
-
ChainForge: A Visual Toolkit for Prompt Engineering and LLM Hypothesis Testing - (May 2024, 15min) Conference-style Paper presentation video for CHI 2024.
-
Say LOL: Testing LLM Robustness to Prompt Injection Attacks using ChainForge - (Jan 2024, 7min) Video showing how to leverage Chainforge to evaluate GPT-4 robustness against several prompt injection attacks.
-
How to Compare Prompts with ChainForge - (Jan 2024, 9min) Live demo, no cuts, no scripts, of ChainForge to compare between different prompts, using template chaining.
-
LLM Prompt Injection Attacks & Testing Vulnerabilities With ChainForge - (May 2023, 5min, by Cobus Greyling).
-
⛓️🛠️ChainForge: An open-source visual programming Interface for LLM flows and more - (May 2023, 10min, by Cobus Greyling).
Blog posts talking about Chainforge (newest first)
-
Hacker News Trending Post - (May 2023) Stayed on Hacker News Front for 2 days after its publication.
-
Simon Willison blog post - (November 2024)
A good option for running evaluations against prompts. -
Introducing ChainForge: A visual programming environment for prompt engineering - (May 2023, by its creator) Medium Post presenting Chainforge.
Your contributions are always welcome! Please take a look at the contribution guidelines first.
This file was validated according to awesome-lint.
If you have any question about this opinionated list, do not hesitate to contact me or open an issue on GitHub.