Interested in learning how to run RL at scale? Here are the best resources to read…

Research on Scaling RL

1. The Art of Scaling RL compute for LLMs: arxiv.org/abs/2510.13786

2. Scaling Behaviors of LLM RL Post-Training: arxiv.org/abs/2509.25300

3. Optimally Scaling Sampling Compute for LLM RL: arxiv.org/abs/2603.12151

4. Scaling up RL: arxiv.org/abs/2507.12507

5. ProRL V2 - Prolonged Training Validates RL Scaling Laws: hijkzzz.notion.site/pro…

6. Polaris - A Recipe for Scaling RL with Reasoning Models: hkunlp.github.io/blog/2…

RL Frameworks

1. Hybrid Flow (outline of the verl framework): arxiv.org/abs/2409.19256

   1. More up-to-date info can be found here: arxiv.org/abs/2601.18150

2. AReal - Large-Scale Async RL: arxiv.org/abs/2505.24298

3. PipelineRL - Fast On-Policy RL: arxiv.org/abs/2509.19128

4. AsyncFlow - Async Streaming RL: arxiv.org/abs/2507.01663

RL for Agents

1. DeepSWE - Open Coding Agent Trained w/ RL: together.ai/blog/deepswe

2. AutoForge - Environment Synthesis for Agentic RL: arxiv.org/abs/2512.22857

3. Agent-R1 - Training Agents w/ End-to-End RL: arxiv.org/abs/2511.14460

4. AgentRL - Scaling RL for Multi-Turn, Multi-Task Agents: arxiv.org/abs/2510.04206

5. The Landscape of Agentic RL: arxiv.org/abs/2509.02547

6. Training SWE Agents with RL: arxiv.org/abs/2508.03501

Case Studies & Tech Reports

1. Kimi tech reports:

   1. Kimi K2 - Open Agentic Intelligence: arxiv.org/abs/2507.20534

   2. Kimi End-to-end Agentic RL: moonshotai.github.io/Ki…

   3. Kimi K1.5 - Scaling RL for LLMs: arxiv.org/abs/2501.12599

2. Composer series from Cursor:

   1. Composer 2: arxiv.org/abs/2603.24477

   2. Composer 2.5: cursor.com/blog/compose…

3. Olmo 3 (also has open code / data): arxiv.org/abs/2512.13961

4. MiniMax tech reports:

   1. MiniMax-M2: arxiv.org/abs/2605.26494

   2. MiniMax-M1: arxiv.org/abs/2506.13585

5. Nemotron 3 (NVIDIA): arxiv.org/abs/2512.20856