Drilldown: Recipe

DeepSeek V4.1 Flash on 4x 8 GB (Morrowmake) (1) · GLM 5.3 Flash on 4x 8 GB (1) · Qwen3.8 Flash Next on 1x 8 GB (ElectroHobo 3 bpw) (1) · Qwen3.8 Flash Next on 2x 8 GB (ElectroHobo W4A16 TP2) (1) · Qwen3.8 Flash Next on 2x 8 GB (mannydelrio PP2) (1)

DeepSeek-V4.1-Flash-1017 (1) · GLM-5.3-Flash (320B MoE, 18B active) (1) · Qwen3.8-Flash-Next (125B MoE per the recipe) (2) · Qwen3.8-Flash-Next-Uncensored (refusal-removed, from orcarouter) (1)

AutoRound mixed 2/3-bit (klee100, 2.998 effective bpw, 47 GiB on card) plus halt95's FP8 n-gram table in host RAM (1) · AWQ W4A16 g32 (leoncca/Qwen3.8-Flash-Next-Uncensored-AWQ-g32); PLE table converted to INT4 and served from SSD (1) · Hybrid: routed experts EXL3 2.9 bpw (LibertAIDAI), everything else official FP8/BF16; KV BF16 (1) · W4A16 (INT4 weights, group size 128, FP16 activations): canada-quant/GLM-5.3-Flash-W4A16-MTP (1) · W4A16: halt95/Qwen3.8-Flash-Next-W4A16-Merlin (AutoRound INT4 g128 experts, INT8 GDN projections, FP8 n-gram table) (1)

Morrowmake/vllm-cmp170hx (vLLM fork, container image) (2) · Morrowmake/vllm-cmp170hx image (Qwen EXL3 release) with patch files mounted read-only (1) · Morrowmake/vllm-cmp170hx image with patch files mounted read-only (1) · wtdcode/vllm-backport 3142ded1 plus a two-file patch and two tinybeast PLE plugins (venv, no container) (1)

recipe 0.2.1; engine 18305948ac (1) · recipe 0.3.1; engine 18305948ac (1) · recipe 1.0.0, engine a0c57b5237 (1) · recipe 1.8.0, engine 819146023f (1) · vllm-backport 3142ded1 (1)

8 GB · Other · None

1 (1) · 2 (2) · 4 (2)

PP2 (layer partition 25,23), no P2P (1) · single card (1) · TP2 with whole-expert expert parallel (1) · TP4 (default) or PP4 (1) · TP4 with expert parallel (96 whole experts per card) (1)

262,144 · 1,048,576 · None · 1,000,000 (YaRN x4 from 262,144) · 1,048,576 (YaRN x4) · 1,048,576 FP16 · 1,048,576 per request · 262,144 (256K in the post) · 262,144 (up to 1,048,576) · 262,144 (weights address 1,048,576) · 262,144 advertised (tested through 32K) · 262,144 native (YaRN to about 500k in the recipe's long variant) · 392,960 (profile q8_384k) · 392,960 per request; 655,360-token Q8 KV pool shared by two requests · about 1,000,000 (8-bit KV cache) · Other

177.8 (1) · 287 (1) · 296.4 (1) · 349.7 (1) · 442.1 (1)

3,161 (1) · 3,818 (1) · 4,243 (1) · 4,696 (1) · 7,006 (1)

64 GB, --gpu-memory-utilization 0.94, TB_TORCH_MEM_CAP_GIB=61.5 (1) · 64 GiB exposed, --gpu-memory-utilization 0.95 (2) · 64 GiB, --gpu-memory-utilization 0.95 (fixed) (1) · 64 GiB, GPU_UTIL 0.95 (fixed) (1)

180 W (5)

580 or newer (measured on 610.57.04) (1) · 580 or newer (measured on 610.57.04); open modules with HMM for the table-on-disk mode (1) · NVIDIA 580 or newer; P2P needs the Morrowmake/cmpunlocker build (2) · nvidia-open 610.43.02 + cmpunlocker v0.5 or newer (1)

cmpunlocker (64 GiB, 74 SMs) (1) · cmpunlocker (64 GiB, 74 SMs) with a P2P-capable build (1) · cmpunlocker (64 GiB, 74 SMs); Morrowmake/cmpunlocker for P2P (2) · cmpunlocker v0.5+ (sudo ./install.sh --profile=8gb --no-iommu) (1)

Proven (2) · Reported (3)

mrmannyatx (1) · nathang6866 (2) · voidgiraffe (2)

October 15, 2026 (1) · October 16, 2026 (1) · October 23, 2026 (2) · October 24, 2026 (1)

https://github.com/ElectroHobo/qwen38-flash-next-3bpw-1x-cmp170hx-recipe (1) · https://github.com/ElectroHobo/qwen38-flash-next-w4a16-2x-cmp170hx-recipe (1) · https://github.com/mannydelrio/qwen38-flash-next-2x-cmp170hx (1) · https://github.com/Morrowmake/dsv41-flash-cmp170hx-recipe (1) · https://github.com/Morrowmake/glm53-flash-cmp170hx-recipe (1)