local intelligence

can it run

Can the AMD Ryzen AI Halo (next-gen, 192GB) run DeepSeek-V4.1-Flash?

STRETCH

Only with compromises. 192GB is below the 212GB working set — expect a lower quant, offloading, or heavy swap. Possible, not pleasant.

~6.1–8.6 tok/s (active-expert estimate)

weights 168.9GB + KV cache 2GB at 4k context vs ~190GB usable → tight. 170.9 of 190 GB usable — keep context modest

computed roofline: 256 GB/s × 0.60–0.85 efficiency window / 25.3GB (Q4_K_M) — bands, never points.

MoE: file is 168.9GB (that decides fit) but only ~82.8B of experts activate per token (~25.3GB streamed) — band below is the active-expert estimate, optimistic; community MoE rows run several× the naive bandwidth math

explore the full catalog84 machines indexed · live prices · what each one can run

Specs

model file168.9GB (GGUF mixed-Q2)
minimum memory212GB working set
this machine192GB unified (160GB VRAM partition)
params552B+196B Engram MoE
licenseMIT
price$5000

Run it

model files: hugging face ↗

easy run: no stock runtime yet

Alternatives

ASUS ExpertCenter Pro ET900N G3 (GB300) — cheapest machine that runs it ($60000)

ASUS ExpertCenter Pro ET900N G3 (GB300) — this model on the biggest machine

Buy

AMD ↗ · Amazon ↗

2026-09-17 · ← full hardware catalog · prices verified at source, may drift