Drilldown: Recipe

DeepSeek V4 Flash 0731 on 3x 8 GB (PP3 DSpark) (1) · DeepSeek V4 Flash 0731 on 4x 8 GB (allover326 PP4 DSpark) (1) · DeepSeek V4.1 Flash heretic on 4x 8 GB (1) · DeepSeek V4.1 Flash on 2x 8 GB (Project Maya) (1) · DeepSeek V4.1 Flash on 4x 8 GB (alanoo fidelity build) (1) · DeepSeek V4.1 Flash on 4x 8 GB (CurtisAccelerate TC-MoE) (1) · DeepSeek V4.1 Flash on 4x 8 GB plus RTX PRO 6000 (ExLlamaV3 TP5) (1) · GLM 5.3 Flash on 2x 8 GB (ExLlamaV3 layer split) (1) · GLM 5.3 Flash on 2x 8 GB (ExLlamaV3 TP2) (1) · GLM 5.3 Flash on 2x 8 GB (Project Maya) (1) · GLM 5.3 Flash on 4x 8 GB (kaka86mm EXL3 + Marlin sidecar + DFlash2) (1) · Qwen3.8 Flash Next on 2x 8 GB (DUFRelic PP2) (1) · Qwen3.8 Flash Next on 2x 8 GB (mannydelrio PP2) (1) · Qwen3.8 Flash Next on 2x 8 GB (Morrowmake EXL3) (1) · Qwen3.8 Flash Next on 2x 8 GB (Plean PP2) (1) · Qwen3.8 Flash Next on 2x 8 GB (Qwen-Flash-SM80, GDS TEP2) (1) · Qwen3.8 Flash Next on 2x 8 GB (vektorprime PP2) (1) · Qwen3.8 Flash Next on 3x 8 GB (clee PP3) (1) · Qwen3.8 Flash Next on 3x or 4x 8 GB (zebgop-ops FP8 PP) (1) · Qwen3.8 Flash Next on 4x 8 GB (1)

DeepSeek-V4-Flash-0731 (2) · DeepSeek-V4.1-Flash-1017 (3) · DeepSeek-V4.1-Flash-1017 (748B in the Maya README) (1) · DeepSeek-V4.1-Flash-1017 heretic (pqhaz abliteration) (1) · GLM-5.3-Flash (320B MoE, 18B active) (2) · GLM-5.3-Flash (320B MoE, 18B active, multimodal) (1) · GLM-5.3-Flash (321B in the Maya README) (1) · Qwen/Qwen3.8-Flash-Next-FP8 (revision bcd9f01d) (1) · Qwen3.8-Flash-Next (6) · Qwen3.8-Flash-Next (nvidia/Qwen3.8-Flash-Next-NVFP4, revision fc694b54, 11 shards, 132.68 GB) (1) · Qwen3.8-Flash-Next-Uncensored (refusal-removed, from orcarouter) (1)

AWQ INT4: cyankiwi/Qwen3.8-Flash-Next-AWQ-INT4; 95 GiB BF16 PLE table in host RAM (FP8 table optional) (1) · AWQ W4A16 g32 (leoncca/Qwen3.8-Flash-Next-Uncensored-AWQ-g32); PLE table converted to INT4 and served from SSD (1) · AWQ W4A16: wtdcode/Qwen3.8-Flash-Next-AWQ-W4A16 (1) · EXL3 3.05 bpw: turboderp/GLM-5.3-Flash-exl3 (1) · EXL3 3.05 bpw: turboderp/GLM-5.3-Flash-exl3; DFlash2 drafter converted to EXL3 6 bpw (1) · EXL3 3.5 bpw (LibertAIDAI; routed experts 4 bpw on layers 0-9 and 30-39, 3 bpw on 10-29; attention and shared experts 6 bpw) plus the official Engram shards (1) · EXL3 4 bpw (Mia-AiLab/GLM-5.3-Flash-EXL3-TR3-4bpw, 164 GB) plus an offline-generated Marlin INT4 decode sidecar (151 GB) (1) · EXL3 5.05 bpw; PLE (n-gram) table read from SSD (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) · Hybrid: routed experts EXL3 2.9 bpw (LibertAIDAI), everything else official FP8/BF16 with the 38 heretic attn.wo_b tensors (1) · Maya GGUF (Maya-S 340 GB or Maya-XS 277 GB; Engram tables stay on SSD) (1) · Maya-M GGUF (116 GB), every expert in VRAM; Maya-S (96.5 GB) and Maya-S24 (94.7 GB) also reported (1) · Mixed INT4 AutoRound (Minachist/Qwen3.8-Flash-Next-INT4-Mixed-AutoRound, about 65 GB) with the full BF16 PLE table on NVMe (1) · Native MXFP4 experts + FP8 attention (official deepseek-ai checkpoint, about 155.4 GiB); KV cache fp8 (2) · NVFP4 main weights through Marlin (no native FP4 on sm_80), mixed FP8 and BF16 MTP weights, FP8 PLE table (about 51.2 GB) kept on NVMe (1) · Official block FP8, about 126 GiB including the about 50 GiB FP8 n-gram (PLE) table, which is served from host RAM; KV cache bf16 (the QSA layers refuse fp8) (1) · W4A16 AutoRound: Intel/Qwen3.8-Flash-Next-W4A16-AutoRound; 95.4 GiB BF16 PLE table streamed from NVMe; KV fp8 e4m3 (software-decoded on SM80) (1) · W4A16 INT4 g128 with the PLE table repacked INT4 g32: zankich/Qwen3.8-Flash-Next-W4A16-Merlin-INT4PLE-FP8KV; KV BF16 (1)

ExLlamaV3 (patched, pinned) (1) · ExLlamaV3 1.5.4 (built for SM80) with a small OpenAI-compatible server (1) · ExLlamaV3 fork (anoane/exllamav3-anemone) + TabbyAPI fork (anoane/tabbyAPI) (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) · Morrowmake/vllm-cmp170hx (EXL3 path) (1) · Morrowmake/vllm-cmp170hx (vLLM fork, container image) (1) · Project Maya (Strata and llama.cpp lineage) (1) · Project Maya (Strata and llama.cpp lineage, EXL3 loading, DFlash2 drafter) (1) · Qwen-Flash-SM80 overlay on vllm/vllm-openai:qwen38-flash-next (digest-pinned), FlashInfer PCIe IPC AllReduce, GDS/cuFile over NVMe P2PDMA (1) · vLLM 0.29.1rc1 nightly plus the checkpoint author's patches (1) · vLLM nightly + PR #53899 (a5530b9) + the repo's patch scripts (1) · vLLM nightly 7d0b4e57 plus the author's pure-Python fork clee/vllm-sm80 (branch q38fn/stack) (1) · vLLM v0.30.0 + zankich INT4-PLE plugin + six hash-guarded patch layers (1) · vLLM: haosdent/vllm branch dsv4-flash-a100 plus allover326's 7 patches (DSpark under PP, sm_80 prefill top-k fallback, logits row chunking) (1) · vLLM: haosdent/vllm branch dsv4-flash-a100 with a vendored copy of allover326's 7 patches (image dsv4-a100:devel) (1) · vLLM: hyd998877/vllm_170hx_glm53-flash-exl3-optimized with kaka86mm's overlay patches; exllamav3 1.4.8 (1) · vLLM: the vllm/vllm-openai:qwen38-flash-next image with zebgop-ops patches 0001 to 0016 (never 0008-WITHDRAWN) and three ported files (1) · wtdcode/vllm-backport (image lazymio/vllm-backport:latest-sm80) (1) · wtdcode/vllm-backport 3142ded1 plus a two-file patch and two tinybeast PLE plugins (venv, no container) (1)

None (2) · 0.1.dev1+ga5530b90c.d20260829 (1) · 0.1.dev20073+g8e685d198; repo commit 2f551bb (patch 0016, 2026-10-02) (1) · base c3046d1 (recommended since 2026-08-13; first released on f8ea5bb); repo commit 3dd2d88 (1) · f8ea5bb-era base with the allover326 patch set as of mid-August 2026 (1) · final production build v30 (numbers measured on v24); repo commit 218b47f (1) · Flun/glm53-flash-cmp170hx-exl3 at 5933112 (1) · GNmac-lab at 2ad5741ab1; image ds41-cmp170hx:1.2.0-fidelity (1) · latest-sm80 tag as of 2026-10-16 (1) · q38fn/stack on nightly-7d0b4e57 (1) · recipe 1.0.0 (commit d4d1f525), engine a0c57b52 (1) · tag qwen38-recipe-1.0.0, engine 18305948ac (1) · tag recipe-09 (309ef6d1); TabbyAPI deepseek-v41 (ce250ff) (1) · upstream dev 0f685aa9fd plus 9 patches; repo db01c256 (1) · v0.3.0 (commit 4d229f1, 2026-10-14); base vLLM 0.1.dev20073+g8e685d198 (1) · v1.0.6 (this row) (1) · vLLM v0.30.0 (ced6857a); plugin zankich/vllm v0.29.0z (6609bfef) (1) · vllm-backport 3142ded1 (1) · vllm-cmp170hx 18305948ac; TC-MoE fb9786ec (1)

8 GB (20)

2 (10) · 3 (2) · 4 (8)

engine-managed expert tiering across VRAM, host RAM and NVMe (1) · engine-managed split with expert tiering across VRAM, RAM and NVMe (1) · layer split (no tensor parallel), no P2P (1) · PP2 (layer partition 25,23), no P2P (1) · PP2 with PLE CPU offload (1) · PP2, MTP 3, no P2P (1) · PP2, MTP depth 3 (1) · PP3 (VLLM_PP_LAYER_PARTITION=15,15,13) with DSpark, 5 draft tokens; PCIe Gen2 x4, no P2P (1) · PP3, no P2P (works at PP2 on two cards) (1) · PP4 (VLLM_PP_LAYER_PARTITION=12,12,12,7) with DSpark, 5 draft tokens; PCIe Gen2 x4, no P2P (1) · PP4 (VLLM_PP_LAYER_PARTITION=13,11,11,10), DFlash2 k=7; PCIe Gen2 x4, no P2P (1) · PP4, VLLM_PP_LAYER_PARTITION=12,12,12,12, MTP 3, FULL_AND_PIECEWISE CUDA graphs; PCIe Gen2 x4 per card, no P2P. Three cards: PP3 with 16,17,15 or 16,16,16 (1) · TEP2 (TP2 plus expert parallel), MTP 6; one card at PCIe Gen2 x16 (x16 mod), one at Gen2 x4; GPU P2P required (1) · TP2 over NCCL (GPU P2P), DFlash2 K7 plus hybrid drafting (1) · TP2 with expert parallel (1) · TP4 (1) · TP4 with expert parallel, MTP 3 (1) · TP4, P2P on (1) · TP4, peer access off (NCCL_P2P_DISABLE=1), host all-reduce (1) · TP5 across four CMP 170HX and one RTX PRO 6000 Blackwell (native backend, reduces through host memory); MTP draft on the PRO (1)

None (2) · 1,000,000 (YaRN x4 from 262,144) (1) · 1,047,736 verified one-shot; 1,002,852 over a 405-turn chat with DSV4_LOGITS_ROW_CHUNK=64 (1) · 1,048,576 (YaRN x4) (1) · 1,048,576 FP16 (1) · 262,144 (7) · 262,144 (256K in the post) (1) · 262,144 (weights address 1,048,576) (1) · 262,144 native; 524,288 with YaRN 2.0; 1,048,576 with YaRN 4.0 (needle verified on PP3) (1) · 392,960 (profile q8_384k) (1) · 392,960 per request; 655,360-token Q8 KV pool shared by two requests (1) · 524,288 (1) · 691,968 (largest stable context at --gpu-memory-utilization 0.965) (1)

60 - 80 (5) · 80 - 100 (5) · 100 - 140 (4) · 140 - 200 (6)

None (4) · 0 - 2,000 (4) · 2,000 - 2,700 (6) · 2,700 - 12,000 (6)

64 GB (4) · 64 GB (65,536 MiB), --gpu-memory-utilization 0.965 (1) · 64 GB, --gpu-memory-utilization 0.85 (1) · 64 GB, --gpu-memory-utilization 0.93 (0.94 is the ceiling) (1) · 64 GB, --gpu-memory-utilization 0.93 (v0.3.0) (1) · 64 GB, --gpu-memory-utilization 0.94, TB_TORCH_MEM_CAP_GIB=61.5 (1) · 64 GiB (64,911 MiB usable), GPU_MEMORY_UTILIZATION 0.98 (1) · 64 GiB, --gpu-memory-utilization 0.90 (1) · 64 GiB, --gpu-memory-utilization 0.92 (miner co-resident) (1) · 64 GiB, --gpu-memory-utilization 0.95 (1) · 64 GiB, gpu-memory-utilization 0.92 (1) · 64 GiB; --gpu-memory-utilization 0.94 plus per-rank KV budgets; measured 34.0-34.3 GiB per 12-layer rank, 37.9 GiB on the drafter rank (1) · 64 GiB; weights PP0 39.48 GiB, PP1 45.53 GiB; --gpu-memory-utilization 0.9 (1) · about 58 GiB of 64 GiB (116.6 GiB resident across both) (1) · about 62.5 GB used of 64 GB, --gpu-memory-utilization 0.97 (1) · about 62.9 / 62.3 GiB idle, peaks about 64.4 / 63.0 GiB with vision (1) · gpu_split 58 GB per CMP, 70 GB on the PRO (1)

None (10) · 100 W (1) · 140 W (1) · 140 W (200 W observed-draw abort guard) (1) · 150 W per CMP (PRO 400 W) (1) · 180 W (2) · 180 W (author's cards); about 170 W with a modest memory OC in his Discord numbers (1) · 200 W (3)

None (12) · 610.43.02 (nvidia-driver-610-open) (1) · 610.43.03 (open modules) (1) · 610.57.04 (1) · NVIDIA 610.43.02 (1) · NVIDIA open 610.43.02 from a cmpunlocker fork with a P2P patch (1) · NVIDIA open 610.43.03, CUDA 13.0 (1) · NVIDIA open 610.57.04 patched with Morrowmake/cmpunlocker 1f2c498 (profile 8gb) (1) · nvidia-open 610.43.02 + cmpunlocker v0.5 or newer (1)

None (10) · amoghmunikote/cmpunlocker 6c442eeb (patch 0016 SM reconfig, 74 SMs) plus Bayley P2P patches 0011/0013/0015 (1) · bayley/cmpunlocker 5a7bb4b, sudo ./install.sh --p2p --no-iommu (1) · cmpunlocker (64 GB) (1) · cmpunlocker (64 GiB); P2P-capable driver for the NCCL backend (1) · cmpunlocker 360acd7 (2026-08-06), 8 GB profile (1) · cmpunlocker fork with P2P (without Morrowmake's TRAP31 patch) (1) · cmpunlocker v0.5+ (sudo ./install.sh --profile=8gb --no-iommu) (1) · cmpunlocker VRAM unlock (64 GiB per card) (1) · cmpunlocker VRAM unlock (65,536 MiB per card) (1) · Morrowmake/cmpunlocker 1f2c498; modprobe RmForceEnableGen2=1, RMPcieLinkSpeed=0x1, ForceP2P=0x11 (1)

Proven (3) · Reported (17)

68x0 (1) · alanoo (1) · anoane (1) · c133 (1) · CurtisAccelerate (1) · darkmaniac7 (1) · deejay9983 (1) · dkpoulsen (1) · dufrelic (1) · flipnburner (1) · matstatman (1) · mrmannyatx (1) · nestaa51 (1) · nguyenthimy2022kg-alt (1) · pcsv. (1) · plean. (1) · tim-odonnell (1) · vektor_prime (1) · voidgiraffe (1) · ZackO2o (1)

August 2026 (3) · September 2026 (3) · October 2026 (13)

https://discord.com/channels/1518634379604136178/1528826330454429706/1560602001908170753 (1) · https://github.com/alanoo81/deepseek-v41-flash-4x-cmp170hx (1) · https://github.com/alexvnesta/glm53-flash-cmp170hx-tp (1) · https://github.com/allover326/deepseek-v4-cmp170hx (1) · https://github.com/anoane/09-deepseek-v41-flash-1017-exl3-tp5-1m (1) · https://github.com/clee/qwen3.8-flash-next-3x170hx (1) · https://github.com/CurtisAccelerate/DeepSeekv41-Flash-1017-EXL3-CMP170HX (1) · https://github.com/DUFRelic/qwen38-flash-next-recipe (1) · https://github.com/Flun/glm53-flash-cmp170hx-exl3 (1) · https://github.com/kaka86mm/glm53-flash-170hx (1) · https://github.com/mannydelrio/qwen38-flash-next-2x-cmp170hx (1) · https://github.com/Morrowmake/dsv41-flash-cmp170hx-recipe (1) · https://github.com/Morrowmake/qwen38-flash-next-cmp170hx-recipe (1) · https://github.com/mw00/project-maya#how-fast-is-it (2) · https://github.com/nguyenthimy2022kg-alt/Qwen-Flash-SM80-170HX (1) · https://github.com/tim-odonnell/cmp170hx-deepseek-v4-flash (1) · https://github.com/vektorprime/qwen38-flash-next-pp2 (1) · https://github.com/zebgop-ops/qwen38-flashnext-pp (1) · https://www.bozocentral.com/2x-cmp170hx-qwen3-flash-next (1)