Drilldown: Recipe

DeepSeek V4.1 Flash heretic on 4x 8 GB · DeepSeek V4.1 Flash on 2x 8 GB (Project Maya) · DeepSeek V4.1 Flash on 4x 8 GB (alanoo fidelity build) · DeepSeek V4.1 Flash on 4x 8 GB (CurtisAccelerate TC-MoE) · DeepSeek V4.1 Flash on 4x 8 GB (GNmac) · DeepSeek V4.1 Flash on 4x 8 GB (Morrowmake) · DeepSeek V4.1 Flash on 4x 8 GB plus RTX PRO 6000 (ExLlamaV3 TP5) · DeepSeek V4.1 Flash on 6x 8 GB (Deepinfer-SM80 TP2 x PP3) · DeepSeek V4.1 Flash on 6x 8 GB (TP6) · DeepSeek V4.1 Flash on 8x 8 GB (Deepinfer-SM80 TP4 x PP2) · DeepSeek V4.1 Flash on 8x 8 GB (Zanooda PP8) · GLM 5.3 Flash on 2x 8 GB (ExLlamaV3 layer split) · GLM 5.3 Flash on 2x 8 GB (ExLlamaV3 TP2) · GLM 5.3 Flash on 2x 8 GB (Project Maya) · GLM 5.3 Flash on 3x 8 GB · GLM 5.3 Flash on 4x 8 GB · GLM 5.3 on 8x 8 GB (Int4-Int8Mix PP8) · GLM 5.3 on 8x 8 GB (vllm-170hx-glm5) · MiMo V2.6 Flash on 4x 8 GB · Qwen3.8 27B on 1x 8 GB (W4A16 vLLM) · Qwen3.8 Flash Next on 1x 8 GB (ElectroHobo 3 bpw) · Qwen3.8 Flash Next on 1x 8 GB (iIIusi0n SSD PLE) · Qwen3.8 Flash Next on 2x 8 GB (DUFRelic PP2) · Qwen3.8 Flash Next on 2x 8 GB (ElectroHobo W4A16 TP2) · Qwen3.8 Flash Next on 2x 8 GB (mannydelrio PP2) · Qwen3.8 Flash Next on 2x 8 GB (Morrowmake EXL3) · Qwen3.8 Flash Next on 2x 8 GB (Plean PP2) · Qwen3.8 Flash Next on 2x 8 GB (vektorprime PP2) · Qwen3.8 Flash Next on 3x 8 GB (clee PP3) · Qwen3.8 Flash Next on 4x 8 GB · Other · None

DeepSeek-V4.1-Flash-1017 (2)

Hybrid: routed experts EXL3 (mostly 3 bpw, 2 bpw on layers 18-22) from LibertAI's 2.9 bpw pack; attention, shared experts, indexer and DSpark drafter official FP8; KV fp8_ds_mla (1) · Hybrid: routed experts EXL3 2.9 bpw (LibertAIDAI), everything else official FP8/BF16; KV BF16 (1)

Morrowmake/vllm-cmp170hx (vLLM fork, container image) (1) · vLLM with a pinned sm_80 patch stack (built locally from source) (1)

recipe 1.0.0, engine a0c57b5237 (1) · recipe 1.2.0 (1)

8 GB (2)

4 (2)

TP4 (default) or PP4 for x1/x4 links (1) · TP4 with expert parallel (96 whole experts per card) (1)

262,144 (1) · 262,144 (up to 1,048,576) (1)

258 (1) · 287 (1)

2,416 (1) · 3,818 (1)

64 GiB exposed, --gpu-memory-utilization 0.95 (1) · 64 GiB, GPU_MEMORY_UTILIZATION 0.95 (1)

180 W (1) · 180 W (nvidia-smi -pl 180) (1)

CUDA 13 capable; measured on 610.43.02 (1) · NVIDIA 580 or newer; P2P needs the Morrowmake/cmpunlocker build (1)

cmpunlocker (64 GiB, 74 SMs) with a P2P-capable driver for TP4 (1) · cmpunlocker (64 GiB, 74 SMs); Morrowmake/cmpunlocker for P2P (1)

Proven (1) · Reported (1)

greencandle9753 (1) · voidgiraffe (1)

October 24, 2026 (1) · October 26, 2026 (1)

https://github.com/GNmac-lab/deepseek-v41-flash-1017-cmp170hx (1) · https://github.com/Morrowmake/dsv41-flash-cmp170hx-recipe (1)