#reinforcement-learning

5 posts

Large-Scale Sharded Weight Transfer with Ray Direct Transport (RDT) in vLLM

Aug 22, 2026·18 min read

We implement a native sharded weight transfer engine in vLLM utilizing Ray Direct Transport (RDT), achieving weight transfer for the Kimi K2 model in BF16 on 48 8xH100 nodes in 7.53s

IsoExec: Unified Execution to Eliminate Trainer-Inference Mismatch in SkyRL

Aug 21, 2026·11 min read

IsoExec unifies numerical execution across SkyRL's vLLM and Megatron runtimes, reducing the average rollout-versus-training logprob difference below 1e-6 on Qwen3.5-35B-A3B with 25% overhead.

vime + ROCm: End-to-End RL Post-Training on AMD Instinct™ GPUs

Jul 10, 2026·10 min read

Announcing ROCm support for vime, now running end-to-end on AMD Instinct MI355X GPUs with prebuilt container.

Announcing vime: A Simple, Stable, and Efficient RL Framework for LLMs

Jun 9, 2026·6 min read

vime connects slime's training stack with vLLM rollouts to provide a simple, stable, and efficient RL post-training pipeline.

Native RL APIs in vLLM

May 28, 2026·12 min read

How vLLM native RL APIs standardize weight syncing and asynchronous RL serving with NCCL and CUDA IPC transfer backends, pause mode, and fixes for fragile DPEP and disaggregated rollout deployments.