Jump to content
Toggle search
Toggle menu
CMP Wiki
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Drilldown: Recipe

More actions
Recipe > cards: 8 GB & power cap : 140 W or 180 W (author's cards); about 170 W with a modest memory OC in his Discord numbers

Use the filters below to narrow your results.

page:
DeepSeek V4.1 Flash on 4x 8 GB (CurtisAccelerate TC-MoE) (1)
model:
DeepSeek-V4.1-Flash-1017 (1)
quant:
Hybrid: LibertAI EXL3 2.9 bpw routed experts (layers 0-39), official FP8/FP4 for attention, shared experts, head, draft and Engram; KV fp8_ds_mla (1)
engine:
Morrowmake/vllm-cmp170hx (bind-mounted sm80 package) plus the author's tensor-core MoE kernels (1)
engine version:
vllm-cmp170hx 18305948ac; TC-MoE fb9786ec (1)
cards: (Click arrow to add another value)
card count:
4 (1)
topology:
TP4, peer access off (NCCL_P2P_DISABLE=1), host all-reduce (1)
context:
262,144 (weights address 1,048,576) (1)
decode tps:
144.2 (1)
prefill tps:
1,523 (1)
vram per card:
about 62.5 GB used of 64 GB, --gpu-memory-utilization 0.97 (1)
power cap: (Click arrow to add another value)
driver:
None (1)
unlock build:
None (1)
status:
Reported (1)
tested by:
CurtisAccelerate (1)
tested on:
October 22, 2026 (1)
source:
https://github.com/CurtisAccelerate/DeepSeekv41-Flash-1017-EXL3-CMP170HX (1)
Error: unclosed string literal.

There are no results for this report.

Retrieved from "https://cmp.wiki/wiki/Special:Drilldown/Recipe"