A Minimal Agent for Automated Theorem Proving

LemmaBench: A Live, Research-Level Benchmark to Evaluate LLM Capabilities in Mathematics (https://arxiv.org/html/2602.24173v1)

Accelerated minigrid environments with JAX

Claude CLI integration for Vim, enabling AI-powered coding workflows.

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

Code for the paper: Linear Control of Test Awareness Reveals Differential Compliance in Reasoning Models

Lean math proofs generated by AlphaProof Nexus and accompanying natural language prose proofs.

DiffusionBlocks: Block-wise Neural Network Training via Diffusion Interpretation

LLM inference in C/C++

SKILL.md for creating animations with Manim. Coding Agent autonomously plans scenes, writes Manim code, renders videos, and refines based on feedback.

utilities for decoding deep representations (like sentence embeddings) back to text

🔖 The Ultimate Markdown Cheatsheet

Browser Harness | Self-healing harness that enables LLMs to complete any task.

Out-of-the-box DeepSeek OCR document parsing Web Studio

A minimal implementation of DeepMind's Genie world model

A script for Google Sheets that connects a FitBit account to a Google Sheet. GPL-3.0-or-later

A script to fetch data from Fitbit servers using their API and store the data in a local influxdb database and visualize them with Grafana

Just enough Kubernetes for you to fly

[ICLR 2026] Tina: Tiny Reasoning Models via LoRA

PeRL: Parameter-Efficient Reinforcement Learning

Testing baseline LLMs performance across various models

The Abstraction and Reasoning Corpus