A project to improve skills of large language models

Democratizing Reinforcement Learning for LLMs

TokenSkip: Controllable Chain-of-Thought Compression in LLMs

[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

📦 Repomix (formerly Repopack) is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Language Models (LLMs) o…

STAR: Spatial-Temporal Augmentation with Text-to-Video Models for Real-World Video Super-Resolution

AI reads books: Page-by-Page PDF Knowledge Extractor & Summarizer. script performs an intelligent page-by-page analysis of PDF books, methodically extracting knowledge points and generating progres…

Official implementation of SVFR.

🎨 Diagram as Code for prototyping cloud system architectures

D2 is a modern diagram scripting language that turns text to diagrams.

Open Source Alternative to Vercel, Netlify and Heroku.

Create an AI capable of solving reasoning tasks it has never seen before

Public repository for "The Surprising Effectiveness of Test-Time Training for Abstract Reasoning"

An open-source, next-generation "runc" that empowers rootless containers to run workloads such as Systemd, Docker, Kubernetes, just like VMs.

MLE-bench is a benchmark for measuring how well AI agents perform at machine learning engineering

A visual and transparent alternative to open-source ChatGPT O1

A minimalist production ready plugin system

An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.

g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains

This project demonstrates a basic chain-of-thought interaction with any LLM (Large Language Model)

A language model programming framework.

DSPy: The framework for programming—not prompting—foundation models

Optimizing inference proxy for LLMs

Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory

Modern Commerce with Next.js and Stripe as the backend.

🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.

Orchestrate zero-shot computer vision models