Agentic reinforcement learning (RL) disaggregates training from rollout, so each policy update must reach the rollout clusters before the next batch. Transferring a full 1T checkpoint for such weight synchronization (refit) takes 87.5 min between two AWS regions. Measurements of BF16 training show t
Llama 3.3 70B on Consumer GPU: The VRAM Calculation Math and KV-Cache Allocation Formula