Metraipolis › archive › dc_315fb2b68c42

Miles v0.1: Production-Level Post-Training — arXiv:2609.08368

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.08368
License (as archived)
CC0-1.0
Archived
2026-10-16T12:58:05.317Z
SHA-256 of the archived text
70db834701f936d3132bb869b406ffb0bf1446a36ad206641331c3d94d717fd4
Stable URL of this page
https://metraipolis.com/documento/dc_315fb2b68c42

How to cite

Miles v0.1: Production-Level Post-Training — arXiv:2609.08368 — arXiv (en), https://arxiv.org/abs/2609.08368 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_315fb2b68c42 (Metraipolis archive dc_315fb2b68c42, sha256 70db834701f936d3132bb869b406ffb0bf1446a36ad206641331c3d94d717fd4)
BibTeX
@misc{dc315fb2b68c42,
  title = {Miles v0.1: Production-Level Post-Training — arXiv:2609.08368},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.08368},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_315fb2b68c42; sha256 70db834701f936d3132bb869b406ffb0bf1446a36ad206641331c3d94d717fd4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.08368. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present Miles v0.1, a full-stack, production-ready system for frontier post-training. Building upon the clean design of slime, Miles designs each stage of the reinforcement-learning (RL) training loop around a single principle: components should be verified, clean, and customizable. With accuracy, efficiency, reliability, and scalability as first-class goals, Miles aims to make frontier-scale RL accessible to researchers and enterprises alike. This report walks through the system end to end: rollout engines built on SGLang, a trainer with a choice of two backends (NVIDIA Megatron-LM and PyTorch FSDP), and three weight-synchronization transports for different deployment topologies. Beyond full-parameter RL, Miles also supports LoRA RL, on-policy distillation, supervised fine-tuning, and true-on-policy rollout-training alignment, and extends the same architecture to diffusion models. We close with an end-to-end case study: fully asynchronous agentic RL on a GLM-5.2 744B-A40B model over terminal-use coding tasks, running on 64 NVIDIA GB300 GPUs with a median step time of 263 seconds over the first 30 measured steps. Miles is open-sourced at this https URL , with the project website at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.08368 — 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 2026-10-16T12:30:25.352Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).