Can tensorrt-llm or how tensorrt-llm support that seprating the prefill stage and decode stage in different GPU or different nodes with self configuration

[GGBond8488]

such as vllm-project/vllm#2809 and https://github.com/LLMServe/DistServe that had done

reference:https://arxiv.org/pdf/2311.18677

[litaotju]

DistServe is on our roadmap, and will be supported in the following versions

[GGBond8488]

DistServe is on our roadmap, and will be supported in the following versions

Thank you for your reply！ I look forward to the future implementation of this feature, which will be a great improvement！

[github-actions]

This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 15 days."