Back to Tools Directory

harness-1

Python

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

Categories & Topics

search-agentsmodel-traininglong-horizonstateful-harness
View on GitHub

Repository Stats

955
GitHub Stars
149
Forks
955
Watchers
1
Open Issues
License
Apache License 2.0
Repository
pat-jj/harness-1
Last Updated
Today
August 7, 2026
Top Contributors
pat-jj
23
zhiyiscs
1
View all contributors →

Related Tools

AlayaWorld

843

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

deep-swe

1.3k

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

Nemotron

1.8k

Developer Asset Hub for NVIDIA Nemotron — A one-stop resource for training recipes, usage cookbooks, datasets, and full end-to-end reference examples to build with Nemotron models

PaperGuru-Benchmark

1.3k

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

RealReplicaBench

1.0k

RealReplicaBench: Benchmarking Long-Horizon Agents in High-Fidelity, Stateful, and Reproducible Replicas of Real Online Services