Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞

The Missing Semester of Your CS Education 📚

Train GPT-2 from scratch on a single GPU. Companion code for the Pre-Training from Scratch book.

Companion code for The Physics of LLM Inference book

Code companion for the RL Post-Training Handbook - training reasoning models on a single GPU

A Cybersecurity Generalist LLM (ICLR'26)

🔥 [ICLR 2025] Official Benchmark Toolkits for "Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark"

Official repo for EMNLP'24 paper "SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning"

Train LLM from scratch for $5 USD - Research.

🔥🔥🔥[AAAI 2026 Oral] Official Implementation of Robust-R1: Degradation-Aware Reasoning for Robust Visual Understanding

PICABench: How Far Are We from Physically Realistic Image Editing?

This is the official repository of UltraHR-100K.

[ICCV 2025 Highlight] OminiControl: Minimal and Universal Control for Diffusion Transformer

Subjects200K dataset

[CVPR 2024] Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers

Visual Generation Tuning

Aligning LMMs with Factually Augmented RLHF

Continuous Thought Machines, because thought takes time and reasoning is a process.

Github repository for "Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging" (ICML 2025)

Github repository for "Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas" (ICML 2025)

Official repo for StableLLAVA

Quick illustration of how one can easily read books together with LLMs. It's great and I highly recommend it.

Awesome Visual Tokenizers/Autoencoders

PatchEval: A New Benchmark for Evaluating LLMs on Patching Real-World Vulnerabilities