DeepSeek-V4 Flash RL on AMD Instinct MI308X GPUs with an AITER-Backed RL Stack

We demonstrate end-to-end DeepSeek-V4 Flash reinforcement learning with an AITER-backed AMD RL stack on 24 AMD Instinct MI308X GPUs. Our test covers the complete workflow on three MI308X OAM nodes—not only inference or an isolated training step, but long-context FP8 rollout, BF16 MoE actor training, online RDMA weight updates, evaluation, and checkpoint recovery. Training and inference run as disaggregated services: two MI308X nodes host a BF16 actor that preserves the numerical stability required by gradient computation and optimizer updates, while a dedicated third MI308X node hosts an FP8 rollout engine whose smaller per-parameter footprint leaves more GPU memory available for KV cache and longer generated sequences. Ray places and coordinates both GPU groups. This experiment demonstrates one tested three-node configuration for running both rollout inference and RL training on MI308X GPUs.