Search

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

An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.

Official implementation of EVOKE: Endless Interactive World with Bounded State and Long-Horizon Supervision. A three-step, CFG-free interactive world model. SOTA on WBench.

a best-effort approximation of a synthetic entity complete with all of the primatives required for long-horizon persistance. batteries included.

Vibe Check is a tool that provides mentor-like feedback to AI Agents, preventing tunnel-vision, over-engineering and reasoning lock-in for complex and long-horizon agent workflows. KISS your over-eager AI Agents goodbye! Effective for: Coding, Ambiguous Tasks, High-Risk tasks

eLLM: Run Long-Horizon Inference Faster on CPUs Than on GPUs

[ICLR 2026] The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Execution

Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents

Runtime for long-horizon agents

Uni-Agent is a framework for training long-horizon agents.

[ICML 2026] Multimodal deep-research MLLM and benchmark. The first long-horizon multimodal deep-research MLLM, extending the number of reasoning turns to dozens and the number of search-engine interactions to hundreds.

SWE-Marathon: an ultra long-horizon SWE benchmark

High-performance AI agent for long-horizon tasks and local models. Built on empirical research. 200k+ tokens of work inside a 64k context window. Run a full dev team in 5GB VRAM.

Programmatic memory for long-horizon LLM agents: the harness appends everything to one log, and the agent searches it with code. 97.4% on ARC-AGI-3 (arXiv:2607.20064)

A living memory system that ingests long-horizon data to infer insights, enabling more decisive action, all while running on a single SQLite file locally.

CheetahClaws: A Fast and Easy-to-Use Agent Harness Infrastructure for Long-Horizon, Multi-Model, and Tool-Using AI Systems

Lifecycle-Aware Memory for long-horizon LLM agents — 66.05% on PaperBench, 94.66% on SurveyBench, 10 peer-reviewed acceptances at FSE/ICML/TOSEM/AEI/ICoGB

Measuring frontier coding agents on original, long-horizon engineering tasks

🚀 Ultra Recipe for Training Long-Horizon Search Agents - matching frontier AI's search capability with a 20B model

🔍 The hardest search benchmark in the wild — vague, multi-turn, proactive. 200 long-horizon tasks with persona-driven progressive disclosure, scored by verifiable schema-free knowledge-graph evaluation. No vibes, just triplet F1.

A control plane with a durable state kernel for long-horizon agents and teams. Keep work moving and improving across sessions, with less human attention.

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

Across 348 long-horizon benchmark sessions, Tura used up to 83.1% fewer turns on the rewrite benchmark and improved the DeepSWE pass rate by up to 16.7 percentage points compared with Codex CLI.

Full-stack open-source interactive long-horizon world model.

The roadmap of long-horizon agents

Cayu is an open-source Python framework for building and running domain-specific long-horizon agents, with a durable runtime built in.

J-Space Cognition Suite — a model-agnostic inference-time control suite for deep reasoning, long-horizon work, verification, and recovery. Based on Anthropic's J-space global workspace research.

AxisAgentic: An Extensible Runtime and Trajectory-Collection Framework for Long-Horizon Agents.

The long-horizon computer-use harness. Run AI agents across desktop apps and the CLI for extended periods while preserving task state and making reliable progress on complex workflows. Features fresh-context execution, durable verified state, independent auditing, recoverable progress, and native Claude Code / Codex / OpenClaw integration.

Open-source harness for long-horizon agent work, with inspectable results and feedback-driven expert squads.