specific method · filed under inference & serving
Encoder-Prefill-Decode disaggregation
Separates encoding, prefill, and decoding so these stages can scale independently.
- sources
- 2
- models
- 2
- labs adopt it
- 2
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 2
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
using a custom SGLang-based engine that disaggregates encoding, prefill and decoding
usedinference servingin GLM-5.3-FlashZ.ai
we adopt Encoder–Prefill–Decode (EPD) disaggregation, enabling vision encoding, prefill, and decoding to scale independently and overlap in execution
usedunclearin DeepSeek-V4.1-FlashDeepSeek
Filed alongside
Other methods under inference & serving :: serving parallelism.
Attention data parallelismDeepEPPrefill-decode disaggregationTensor parallelism (degree 4)Tensor parallelism (degree 8)Topology-aware NVLink domain placementZero-copy fused token permutation and unpermutationData-parallel vision encodingExpert parallelismFused reduce-scatter/all-gather collectivesIdentical cache-layout pinning across prefill and decode poolsLow-precision MoE combineRound-robin routing for prefill-decode disaggregationTensor parallelism for MoE layersTensorRT-LLM all-reduce backendToken migration for balanced expert placement