Tracking how reinforcement learning scales across models, compute, environments, rewards and agents — and helping teams turn frontier research into working systems.
Latest research
Jingxiao Yang
A work-in-progress study of long-horizon agent reinforcement learning that uses privileged process supervision to redistribute verified trajectory-level credit across executable actions.
Ziniu Li
An empirical decomposition of LLM reinforcement-learning batch scaling into sample-indexed learning behavior and systems throughput, with a decision rule for when a larger batch lowers wall-clock time-to-target.
Qiancheng Zhou
A controlled study of where reasoning diversity disappears during RLVR, distinguishing failure to initiate a solution family from failure to execute it after the first reasoning choice.
Xingyu Shen
A method for preserving reasoning diversity during RLVR by conditioning the target model on partial trajectories produced by a smaller, weaker model.
Maciej Besta
A compute-centric systematization of RL-style reasoning-model post-training, covering PPO and GRPO as well as intra-model, inter-model, synchronous, and asynchronous parallelism.
Zichen Liu
A controlled examination of R1-Zero-style reinforcement learning that separates robust training effects from claims that are sensitive to setup, reward design, and evaluation choices.
Frontier model reports
Z.ai
An official GLM-5.3 release note describing a post-training-only update on the same base model as GLM-5.2, with expanded executable environments for complex coding, security, and long-horizon professional workflows.
Moonshot AI
A technical report on Kimi K3, a 2.8T-parameter sparse multimodal model with 104B active parameters and a one-million-token context window, post-trained with reinforcement learning across general, agentic, coding, and reasoning domains.
Z.ai
An official release of the 750B-A40B GLM-5.2 model, combining a one-million-token context, IndexShare sparse-attention reuse, and larger-scale agentic reinforcement learning for long-horizon tasks.
MiniMax
An official release of MiniMax-M3, a natively multimodal sparse model with about 428B total and 23B active parameters, a one-million-token context window, and agent-oriented coding and cowork capabilities.
Z.ai
An official GLM-5.1 model update focused on keeping an agent productive across longer coding and engineering runs through repeated execution, inspection, diagnosis, and strategy revision.
Z.ai
A GLM-5 technical report centered on agentic engineering, combining a more efficient long-context architecture with asynchronous reinforcement-learning infrastructure and agent RL for complex, long-horizon software tasks.
Featured analysis
A working framework
A practical framework for locating the binding constraint in an RL system across environments, rollout generation, reward quality, optimization, and evaluation.
Read the analysis →Browse by topic
RL Scaling studies what happens when reinforcement learning systems grow across compute, data, model capacity, rollout volume, task diversity, and feedback quality.
→02Reinforcement Learning with Verifiable Rewards uses outcomes that can be checked by programs, tests, simulators, or formal rules.
→03Group Relative Policy Optimization and related methods update a language-model policy using rewards from sampled response groups rather than a separately learned value critic.
→04Agentic RL applies reinforcement learning to systems that plan, use tools, interact with software, and act over long horizons.
→05An RL environment defines the tasks, observations, actions, transitions, and outcomes through which a policy gains experience.
→06Rewards translate outcomes and preferences into the signal used to improve a policy.
→07Post-training turns a pretrained model into a useful system through supervised fine-tuning, preference optimization, reinforcement learning, distillation, and targeted data generation.
→08Scaling laws describe empirical relationships between resources and performance.
→Research into systems
Focused engagements for teams facing a high-value research, evaluation, or engineering decision.
Discuss a project ↗A decision-ready landscape of models, papers, competitors, and technical options for a specific strategic question.
Custom evaluation sets and a quality, cost, latency, and regression baseline for your model or agent system.
Workflow design, implementation, evaluation, observability, and handover for production-oriented AI agents.
RL Scaling is an independent AI research and engineering studio. We publish source-linked research notes, explain why developments matter, and work with teams on frontier research, evaluation, and agent systems.
About the studio →