⌥ AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more

π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.

slime is an LLM post-training framework for RL Scaling.

A powerful AI coding agent. Built for the terminal.

Build, run, and manage agent platforms.

Official Implementation of "KBLaM: Knowledge Base augmented Language Model"

🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation

Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch

Detect and extract tables to markdown and csv

🌎💪 BrowserGym, a Gym environment for web task automation

A programming framework for agentic AI

YC (S26) | Open Computer History | Record your screen continuously locally and provide context to your agents (Claude, Codex, Openclaw, Hermes, Runner...)

Playwright is a framework for Web Testing and Automation. It allows testing Chromium, Firefox and WebKit with a single API.

Code for "WebVoyager: WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models"

An LLM-based Web Navigating Agent (KDD'24)

A collection of papers on the topic of ``Computer Vision in the Wild (CVinW)''

tiny vision language model

Code for the paper 🌳 Tree Search for Language Model Agents

VisualWebArena is a benchmark for multimodal agents.

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without …

ScreenQA dataset was introduced in the "ScreenQA: Large-Scale Question-Answer Pairs over Mobile App Screenshots" paper. It contains ~86K question-answer pairs collected by human annotators for ~35K…

gpt-2 from scratch in mlx

Compile a demonstrated GUI workflow into a deterministic, locally executable program. Zero model calls on healthy runs; governed repair; halts instead of guessing. Launcher for openadapt-flow: pip …

[ICML'24] SeeAct is a system for generalist web agents that autonomously carry out tasks on any given website, with a focus on large multimodal models (LMMs) such as GPT-4V(ision).

Cohere Toolkit is a collection of prebuilt components enabling users to quickly build and deploy RAG applications.

A high-throughput and memory-efficient inference and serving engine for LLMs

mPLUG-DocOwl: Modularized Multimodal Large Language Model for Document Understanding

Convert PDF to markdown + JSON quickly with high accuracy

[Nature Reviews Bioengineering🔥] Application of Large Language Models in Medicine. A curated list of practical guide resources of Medical LLMs (Medical LLMs Tree, Tables, and Papers)