Reference implementation + data for 'Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information'

MobileLLM-R1

LLM-as-a-Verifier is a general-purpose framework that provides fine-grained feedback for any agent without requiring additional training. It achieves SOTA performance across coding, robotics, and m…

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

Vision-OPD is a regional-to-global on-policy self-distillation framework that transfers a model's own privileged crop-conditioned perception to its full-image policy, enabling fine-grained visual u…

🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.

Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.

A framework for few-shot evaluation of language models.

The agent that grows with you

JoyAI-Echo-1.5: Long-Horizon Audio-Visual Generation for Persistent Stories and Interactive Worlds

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

Bernini is a unified framework for video generation and editing that combines an MLLM-based semantic planner with a DiT-based renderer.

Repo for paper "Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability"

🧠 Train a 64M-parameter LLM from scratch in just 2h!

Official repo for "Let ViT Speak: Generative Language-Image Pre-training"

Optimize prompts, code, and more with AI-powered Reflective Optimization

A simple screen parsing tool towards pure vision based GUI agent

Archived — ML Intern is no longer maintained. Continue with HuggingChat.

Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe

An agentic skills framework & software development methodology that works.

A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.

An AI Hedge Fund Team

An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

A curated list of papers, tools, and resources on Multi-Token Prediction (MTP) and related techniques in Large Language Models (LLMs), Speech-Language Models (SLMs), and more.

Teams-first Multi-agent orchestration for Claude Code

Your own professional personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

Visual Causal Flow

The AI that really does things. Any OS. Any Platform. The lobster way. 🦞

JarvisX-Cowork: Your First Personal AI Creative Assistant for Everyone!

A modular graph-based Retrieval-Augmented Generation (RAG) system