LoongSage: A Production-Grade, High-Efficiency Agentic RL Framework for Frontier LLMs with Validated Recipes

loongsage:-a-production-grade,-high-efficiency-agentic-rl-framework-for-frontier-llms-with-validated-recipes

LoongSage: A Production-Grade, High-Efficiency Agentic RL Framework for Frontier LLMs with Validated Recipes

[Submitted on 12 Oct 2026 (v1), last revised 18 Nov 2026 (this version, v3)]

View PDF HTML (experimental)

Abstract:Reinforcement learning (RL) has emerged as a critical phase in the post-training pipeline of large language models (LLMs). As the frontier of LLM capabilities shifts from static reasoning to agentic tasks–characterized by external tool invocation, multi-turn interactions, and long-horizon decision-making–existing mainstream RL frameworks have grown increasingly unwieldy, raising the overhead of adoption, usage, and debugging. To address these challenges, we present LoongSage, an agent-native, fully asynchronous RL framework with a lightweight core of roughly 14K lines of code. Built upon Megatron for distributed training and SGLang for high-throughput inference, LoongSage centralizes orchestration within a single controller process and encapsulates all GPU-resident components as Ray actors. It introduces AgentFlow, a unified execution paradigm that orchestrates both white-box and black-box agents. For asynchronous execution, it integrates sliding-window strategies that bound trajectory staleness, and employs TransferMesh, a topology-aware synchronization mechanism that routes weights after each step over the fastest transport available to each receiver. Beyond agentic RL, LoongSage incorporates a memory-efficient Teacher Manager that decouples the scaling of teacher models from GPU memory constraints, natively supporting Multi-Teacher On-Policy Distillation. Extensive evaluations across four complete training runs, spanning complex reinforcement learning and multi-teacher distillation, demonstrate that task rewards improve steadily while the train-inference gap is kept under continuous control, and the fully asynchronous pipeline removes synchronous idle bubbles, sustaining hardware utilization.

Submission history

From: Minghan Zhu [view email]
[v1] Mon, 12 Oct 2026 08:08:16 UTC (998 KB)
[v2] Fri, 16 Oct 2026 03:07:10 UTC (998 KB)
[v3] Wed, 18 Nov 2026 06:24:06 UTC (770 KB)