Illustration drawn by RepoMasters from the repo’s ID and its language color. Credits
PrototypeCommon

chadhurley25075-png/pd-bridge

Heterogeneous prefill/decode for DeepSeek-V4-Flash: CUDA prefill (DGX Spark, vLLM) -> Metal decode (Mac Studio, oMLX) over plain 10GbE

Owner
chadhurley25075-png (personal account)
Stars
Stars: 151
Forks
Forks: 13
Language
Python
License
Apache-2.0
Created
2026
Topics
apple-silicondeepseekdgx-sparkheterogeneous-inferencekv-cachellm-inferencemlxomlxprefill-decodevllm
Tier
Prototype (100 to 249 ★) What this tier means
View on GitHub

Nearby in Python

All Python repos