ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents

Continual Learning Mechanisms Compose for Long-Horizon Memorization

Just a VLM Agent Can Play Robots 🦾

[ANRnd 병합 대상] Recursive Experiential–Working Memory Evolution for Long-Horizon Agent Harnesses

[ANRnd 병합 대상] A self-improving RLM agent for coding workflows and long-running autonomous tasks.

🧩 FrontierAgent, our agent framework, open-sourced alongside it — native command-line TUI, ReAct and Agent Team modes, one command on macOS and Linux, no preinstall, no hard Docker dependency.

An Omni-Modal, Omni-Discipline AI Scientist

DeepSeek Harness: Everything is a Plugin.

LLMRouter: An Open-Source Library for LLM Routing

RouteProfile: Graph-Based Profiling for Cold-Start LLM Routing

Official code for "SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization"

SimpleMem: Efficient Lifelong Memory for LLM Agents — Text & Multimodal

KV cache compression via block-diagonal rotation. Beats TurboQuant: better PPL (6.91 vs 7.07), 28% faster decode, 5.3x faster prefill, 44x fewer params. Drop-in llama.cpp integration.

Memento-Skills: Let Agents Design Agents

Memento-Skills: Let Agents Design Agents

Personal AI, On Personal Devices

Official repo of "MMBench-GUI: Hierarchical Multi-Platform Evaluation Framework for GUI Agents". It can be used to evaluate a GUI agent with a hierarchical manner across multiple platforms, includi…

Being-H is BeingBeyond's family of human-centric embodied foundation models.

NVIDIA Alpamayo 1 Nano is an open 10B reasoning VLA model for autonomous vehicles that pairs driving trajectories with Chain-of-Causation reasoning.

Extending the Context of Pretrained LLMs by Dropping Their Positional Embedding

PyTorch building blocks for the OLMo ecosystem

Omnilingual ASR Open-Source Multilingual SpeechRecognition for 1600+ Languages

The official GitHub mirror of the Chromium source

Contexts Optical Compression