DeepSeek V4 Flash 284B (13B active)
on GH200 · vLLM · 1,048,576 ctx
Summary
User reports 64 t/s baseline and 171 t/s with DSpark k=5, a 2.7x speedup.
Setup is vLLM nightly 0.27.2rc1.dev122, with 88 GB of experts offloaded to LPDDR5x via UVA.
on GH200 · vLLM · 1,048,576 ctx