# When to Use Encode-Prefill-Decode Disaggregation to Accelerate Multimodal Model Serving — retained record

> Retained news record from news:nvidia-developer-blog.

Canonical: https://highsignal.app/data/records/fd913d9ec8ec12a2

[← sources](https://highsignal.app/data)


source: [news](https://highsignal.app/data/news) (news:nvidia-developer-blog)  published 2026-09-09 20:31 ingested 2026-09-10 02:55

[original source ↗](https://developer.nvidia.com/blog/when-to-use-encode-prefill-decode-disaggregation-to-accelerate-multimodal-model-serving/)

## retained text

No document body was retained for this record — only its metadata was stored.

document fetched 2026-09-10 02:55 · canonical https://developer.nvidia.com/blog/when-to-use-encode-prefill-decode-disaggregation-to-accelerate-multimodal-model-serving