Skip to content
#

offload

Here are 42 public repositories matching this topic...

DGX sPLIce: DeepSeek-V4.1-Flash on 2x DGX Spark + 2x RTX 3090 drafter, x86 offload (Engram over RDMA, DSpark drafter). Prefill 2,606-2,827 tok/s to 256K; 322 tok/s structured, 192 tok/s code at 4 streams; 1.93M-token KV pool. Full build, install guide, raw JSON.

  • Updated Nov 1, 2026
  • Python

Add this topic to your repo

To associate your repository with the offload topic, visit your repo's landing page and select "manage topics."

Learn more