Metraipolis › archive › dc_fbd0624adce6

Agent Lightning v1.0: Towards Harnessed Agentic RL — arXiv:2608.17528

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.17528
License (as archived)
CC0-1.0
Archived
2026-10-18T14:28:42.541Z
SHA-256 of the archived text
316fe0b82eea6bf768d76b561690b773720df58a8872cdffc99e3fc0808de2b6
Stable URL of this page
https://metraipolis.com/documento/dc_fbd0624adce6

How to cite

Agent Lightning v1.0: Towards Harnessed Agentic RL — arXiv:2608.17528 — arXiv (en), https://arxiv.org/abs/2608.17528 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fbd0624adce6 (Metraipolis archive dc_fbd0624adce6, sha256 316fe0b82eea6bf768d76b561690b773720df58a8872cdffc99e3fc0808de2b6)
BibTeX
@misc{dcfbd0624adce6,
  title = {Agent Lightning v1.0: Towards Harnessed Agentic RL — arXiv:2608.17528},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.17528},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fbd0624adce6; sha256 316fe0b82eea6bf768d76b561690b773720df58a8872cdffc99e3fc0808de2b6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.17528. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern agents operate inside agent harnesses that manage tools, context, and control flow, making the harness a critical part of the agent system. Our original Agent Lightning introduced a disaggregated architecture that connects arbitrary agents to RL training through an LLM endpoint proxy, an approach later adopted by frameworks such as verl Uni-Agent, AReaL 2.0, slime, and Polar. We refer to this paradigm as harnessed agentic RL, where the deploy-time harness directly participates in model post-training. Harnessed agentic RL differs fundamentally from traditional agentic RL: the harness, rather than the training engine, owns the environment interaction loop, while the trainer observes only sequences of LLM request-response pairs. This introduces challenges in retokenization, sample merging, advantage calculation, loss normalization, and backend scheduling, which can substantially affect training stability and effectiveness. We present Agent Lightning v1.0, a lightweight framework for harnessed agentic RL implemented in approximately 3,500 lines of code. It supports arbitrary agent harnesses and serves as a practical testbed for studying these challenges. We evaluate it on instruction-following, search, and coding agents, and provide a complete reproducible pipeline for coding-agent RL. Using only 6K training examples and modest compute, RL improves Qwen3.5-9B on SWE-bench Verified from 41.8% to 56.4%, a 14.6-point absolute gain. We release the complete workflow and training scripts to facilitate reproducible research on harnessed agentic RL.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-19 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2608.17528 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T14:27:54.273Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-19?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).