# vLLM cuts DeepSeek V4.1 Flash multi-page document TTFT by 44–47%

> Original title: Building on the existing EPD architecture, our latest improvements make disaggregation more efficient through better data transfer, cache...

- Source: vLLM
- Published: 2026-11-11T10:03:44.000Z
- Bad Influence: https://badinfluence.news/items/dui5o1y6s2yyeks80pgzc1qvf
- Original: https://x.com/vllm_project/status/2120457441729298458

## Summary

vLLM says its latest EPD disaggregation improvements, covering data transfer, cache reuse, and routing, cut P90 time to first token by 44–47% in a DeepSeek V4.1 Flash multi-page document benchmark at concurrency 64–128. The same benchmark shows token throughput per GPU rising 6.6–9.5%, counting every encoder GPU.

## Post

Building on the existing EPD architecture, our latest improvements make disaggregation more efficient through better data transfer, cache reuse, and routing.
 
In our DeepSeek V4.1 Flash multi-page document benchmark at concurrency 64–128: 
⚡️44–47% lower P90 TTFT 
🚀6.6–9.5% higher token throughput per GPU, counting every encoder GPU.