GLM-5: From Vibe Coding to Agentic Engineering

Code that accompanies the paper release for "LLMs Corrupt Your Documents When You Delegate"

SWE-Marathon: an ultra long-horizon SWE benchmark

Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).

🧠 Awesome Memory-VLA: A curated list of Visual-Language-Action models with memory

Official Repository for our paper: PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems

Code for Scalable Offline Model-Based RL with Action chunking

Code for Tackling Long-Horizon Tasks with Model-based Offline Reinforcement Learning

The simplest way to build long-horizon environments

VLM-RL Hierarchical Loco-Manupilation For Long-Horizon Tasks With G1 robot in Isaac Lab/Sim

Local-first, eval-first memory for long-horizon AI agents — no LLM at ingest. Python SDK + MCP server with source-traceable recall, belief revision, selective forgetting, and reproducible benchmarks.

A family of long-horizon software-engineering environments for OpenEnv, adapted from https://github.com/Proximal-Labs/frontier-swe

MobileMem: On-Device Memory for Continually Evolving Agents

Official repository for the paper: Can AI Agents Synthesize Scientific Conclusions?

Pushing the Limits of Laboratory 3D Perception and Long-Horizon Planning via Protocol-Aligned Action Prediction

Steady-Forcing: Balancing Spatial Persistence and Motion Continuity in Long-Horizon Nature Video Diffusion

A real-world inspired environment for selective context retention under noise. It evaluates an LLM's ability to manage a fixed-capacity memory buffer, retaining high-value information while filtering out distractors

OpenClaw humanity infusions OtherPowers Creative Intelligence Agency. 🦞

Dashboard for real-world inspired environment for selective context retention under noise. It evaluates an LLM's ability to manage a fixed-capacity memory buffer, retaining high-value information while filtering out distractors

Portable harness for autonomous, long-horizon, multi-agent workflows