Drilldown: Recipe

DeepSeek V4.1 Flash on 4x 8 GB (alanoo fidelity build) (1) · DeepSeek V4.1 Flash on 4x 8 GB (CurtisAccelerate TC-MoE) (1) · GLM 5.3 on 8x 8 GB (Int4-Int8Mix PP8) (1)

DeepSeek-V4.1-Flash-1017 (2) · GLM-5.3 (full 78-layer, not Flash) (1)

GLM-5.3-Int4-Int8Mix (INT4 / INT8 mix); KV fp8_ds_mla (1) · Hybrid: all 40 expert layers at 3 bits (LibertAI 2.9 bpw pack with layers 18-22 grafted from the 3.5 bpw pack), official FP8 dense tensors (1) · 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)

GNmac-lab recipe image 1.2.0 (vLLM with sm_80 patches) plus three patches (1) · Morrowmake/vllm-cmp170hx (bind-mounted sm80 package) plus the author's tensor-core MoE kernels (1) · vLLM (build not stated) (1)

None (1) · GNmac-lab at 2ad5741ab1; image ds41-cmp170hx:1.2.0-fidelity (1) · vllm-cmp170hx 18305948ac; TC-MoE fb9786ec (1)

8 GB · Other · None

4 (2) · 8 (1)

PP8, TP1 (1) · TP4, P2P on (1) · TP4, peer access off (NCCL_P2P_DISABLE=1), host all-reduce (1)

1,048,576 (1) · 262,144 (1) · 262,144 (weights address 1,048,576) (1)

25 (1) · 105 (1) · 144.2 (1)

None (1) · 1,523 (1) · 2,514 (1)

64 GiB (64,911 MiB usable), GPU_MEMORY_UTILIZATION 0.98 (1) · 64 GiB, --gpu-memory-utilization 0.97 (1) · about 62.5 GB used of 64 GB, --gpu-memory-utilization 0.97 (1)

None · 180 W · 200 W · 140 W · 140 W (200 W observed-draw abort guard) · 150 W per CMP (PRO 400 W) · 170 W (author's 10-card run) · 180 W (author's cards); about 170 W with a modest memory OC in his Discord numbers · 180 W (nvidia-smi -pl 180) · 180 W (one card unstable at 250 W) · Other

None (2) · NVIDIA open 610.43.02 from a cmpunlocker fork with a P2P patch (1)

None (2) · cmpunlocker fork with P2P (without Morrowmake's TRAP31 patch) (1)

Reported (3)

1a2b3c0967 (1) · alanoo (1) · CurtisAccelerate (1)

October 7, 2026 (1) · October 22, 2026 (1) · October 27, 2026 (1)

https://discord.com/channels/1518634379604136178/1528826330454429706/1557523316120748073 (1) · https://github.com/alanoo81/deepseek-v41-flash-4x-cmp170hx (1) · https://github.com/CurtisAccelerate/DeepSeekv41-Flash-1017-EXL3-CMP170HX (1)