cs.LGNov 18, 2026

RSIBench-Infra: Can Agents Optimize Their Own Training and Inference Infrastructure at Production Scale?

Authors: Hanlin Wu, Yifei Wang, Zhicheng Cai, Huan-ang Gao, Wei-Ying Ma, Ya-Qin Zhang, Jingjing Liu, Hao Zhou

Organizations: ByteDance Seed · Institute for AI Industry Research (AIR), Tsinghua University · SIA-Lab of Tsinghua AIR and ByteDance Seed

Abstract

Improving the efficiency of training and inference infrastructure is a practical component of recursive self-improvement. Existing benchmarks for agent-driven optimization largely focus on individual software layers such as kernel, leaving unclear whether agents can deliver end-to-end performance gains under production workloads. We introduce RSIBench-Infra, a benchmark of 20 training and inference tasks spanning diverse model families and modalities, with deployments of up to 32 accelerators on NVIDIA GPUs and Ascend NPUs. Agents start from runnable baselines derived from official recipes and can modify the software stack across engine, kernel, and instruction levels. An OCI image-layer submission protocol captures these changes for reproducible deployment, while independent evaluation measures end-to-end performance under task-specific quality constraints. Across all task results, 66.2% exceed 1.10×1.10\times validated speedup. These results demonstrate that current agents can improve production-scale training and inference performance, but identifying effective optimization strategies and consistently translating local improvements into end-to-end gains remain challenging.

Explore similar work

Nov 8, 2026

InfraRSI: Harnessing Recursive Self-Improvement for Training Infrastructure

While Recursive Self-Improvement (RSI) accelerates isolated algorithm and kernel synthesis, distributed LLM training depends on tightly coupled launch settings, operator kernels, and communication schedules, where single edits often hit out-of-memory (OOM) boundaries, compiler conflicts, measurement jitter, or silent numerical drift. We present InfraRSI, the agentic harness that turns open-ended human commands into verified training optimizations via a five-stage workflow: init locks editable scope and a fixed task contract over workload, loss, and timing invariants to prevent reward hacking, doctor profiles baseline phases and calibrates cluster noise, run co-optimizes launch knobs and code patches before re-verifying top candidates against the baseline and logging findings to a cross-hardware experience bank, and report and apply audit source paths, track failed-trial costs, and commit promoted changes. Rather than stepping greedily from a single incumbent, InfraRSI initializes from diverse experience-seeded configurations and fits global execution-time and censored peak-memory models across all trials, including OOMs. Using these models, it prices per-setting memory costs, revises miscalibrated LLM priors from live runs, and pairs memory-hungry optimizations with memory savers and verified code patches. Across four single- and multi-node dense and Mixture-of-Experts (MoE) LLM workloads, InfraRSI achieves a speedup of up to 2.75×2.75\times and lifts mean Model FLOPs Utilization (MFU) from 17.4%17.4\% to 30.0%30.0\%, outperforming existing tuners and coding agents while transferring experience across GPUs. As it improves training infrastructure efficiency, InfraRSI forms a two-level Recursive Self-Improvement loop---revising priors online within runs and distilling experience across clusters---to push foundation model scaling.
Nov 9, 2026

Infrathon: Benchmarking Language Agents for Real-World, Long-Horizon AI Infra Optimization

In this work, we study whether frontier agents can continually improve real-world AI infrastructure over long horizons. We introduce Infrathon, a benchmark of diverse, long-horizon AI infrastructure optimization tasks across different levels. We evaluate six models across 150 runs with 24-hour budgets, totaling over 10,000 GPU-hours. Our results show that model rankings vary with the evaluation horizon, while 135 of the 150 runs achieve further improvements during the second half of their budgets. These findings indicate that shorter evaluations can overlook both optimization gains and changes in relative model performance. We also find agents can discover solutions that outperform human-designed baselines: Claude Opus 5.5 increases the mean KV-cache compression ratio from 6.82×6.82\times with the human-designed Quant-VideoGen codec to 19.86×19.86\times. Analysis of agent trajectories identifies task violations, unreliable testing, and premature completion as barriers to sustained improvement. Together, our Infrathon framework provides resources for constructing long-horizon tasks, evaluating frontier agents, and discovering good infrastructure solutions. Code is available at https://github.com/InfrathonX/Infrathon
Oct 23, 2026

RSI-Opt: Recursive Self-Improvement Optimizer Agents For Model Training

Conventional neural-network training typically follows a predefined optimization recipe, motivating a LLM agent that monitors training dynamics and adjusts optimizer settings as training progresses, but a fixed agent policy does not accumulate experience across runs. We introduce RSI-Opt, a plug-and-play Recursive Self-Improvement Optimizer Agents which evolves the knowledge that guides this process: a prompt, a playbook of procedures, a library of tested optimizer code, and a decision cadence. A meta-agent rewrites this policy using evidence from earlier runs, and an archive retains evaluated policies for further evolution. Across 5 generations on three training tasks, the meta-agents develop reusable procedures and code and conduct their own CPU proxy experiments. RSI-Opt substantially improves optimizer-agent performance, with representative raw-space summary zz-scores increasing from +10.48+10.48 to +14.33+14.33 on the evolution suite with Opus 5.5 and from −0.36-0.36 to +3.11+3.11 on held-out tasks with GPT-5.5 through cross-model policy transfer. On the Burgers PDE, it reduces mean relative L2L_2 error by approximately 2.7-fold with Opus 5.5 and 8.8-fold with GPT-6.1 over the fixed-prompt agent baseline, although PDE transfer varies across LLMs and tasks. Controlled comparisons show that the evolved policy combines reusable optimization knowledge with adaptation to each task: initial decisions capture most observed gains on seen tasks, while continued adaptation improves outcomes on unseen tasks. These findings demonstrate how procedures and executable tools accumulated across training runs can improve optimizer-agent performance and support transfer to scientific machine learning.