Drilldown: Recipe
More actions
Use the filters below to narrow your results.
page:
DeepSeek V4.1 Flash on 5x 8 GB (PP5 full residency) (1) ·
GLM 5.3 Flash on 4x 8 GB (kaka86mm EXL3 + Marlin sidecar + DFlash2) (1) ·
GLM 5.3 on 8x 8 GB (vllm-170hx-glm5) (1) ·
Qwen3.8 Flash Next on 1x 8 GB (iIIusi0n SSD PLE) (1) ·
Qwen3.8 Flash Next on 2x 8 GB (Qwen-Flash-SM80, GDS TEP2) (1) ·
Qwen3.8 Flash Next on 3x or 4x 8 GB (zebgop-ops FP8 PP) (1)
model:
DeepSeek-V4.1-Flash (1) ·
GLM-5.3 (full 78-layer DSA MoE, not Flash) (1) ·
GLM-5.3-Flash (320B MoE, 18B active, multimodal) (1) ·
Qwen/Qwen3.8-Flash-Next-FP8 (revision bcd9f01d) (1) ·
Qwen3.8-Flash-Next (1) ·
Qwen3.8-Flash-Next (nvidia/Qwen3.8-Flash-Next-NVFP4, revision fc694b54, 11 shards, 132.68 GB) (1)
quant:
AutoRound 3 bpw + MTP: klee100/Qwen3.8-Flash-Next-AutoRound-3bpw-MTP; 95.37 GiB BF16 PLE table on SSD (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) ·
GLM-5.3-Int4-Int8Mix-AWQ-g64; drafter incoai/GLM-5.3-DFlash2 (1) ·
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) ·
Official checkpoint (510 GB: MXFP4 routed experts, FP8 dense layers, 203 GB FP8 Engram table); KV cache fp8_ds_mla (1)
engine:
Morrowmake/vllm-cmp170hx branch ampere-glm53 plus fail-closed source patches (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 a5a30471 plus a 13-file downstream patch (native Linux AIO PLE reader) (1) ·
vLLM: 344303947's fork of wtdcode/vllm-backport (shadow source layer, staged Marlin repack, exact-size Engram pinning) (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vLLM: the vllm/vllm-openai:qwen38-flash-next image with zebgop-ops patches 0001 to 0016 (never 0008-WITHDRAWN) and three ported files (1)
engine version:
0.1.dev20073+g8e685d198; repo commit 2f551bb (patch 0016, 2026-10-02) (1) ·
0.29.1rc1.dev402+ga5a30471f.ple1 (1) ·
final production build v30 (numbers measured on v24); repo commit 218b47f (1) ·
v0.13.1.dev6, tag v41-pp5-1m-pinmem-20260916 (1) ·
v0.3.0 (commit 4d229f1, 2026-10-14); base vLLM 0.1.dev20073+g8e685d198 (1) ·
vLLM 0.30.1rc1, fork commit 378c37b0 (1)
cards:
topology:
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) ·
PP5 with an even 8,8,8,8,8 layer split; the repo's cards are on PCIe Gen2 x16 with no NVLink (1) ·
PP8 (also PP9 and PP10), TP1 (1) ·
single card (1) ·
TEP2 (TP2 plus expert parallel), MTP 6; one card at PCIe Gen2 x16 (x16 mod), one at Gen2 x4; GPU P2P required (1)
context:
decode tps:
prefill tps:
vram per card:
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, full residency (no expert offload), --gpu-memory-utilization 0.97 (1) ·
64 GiB, --gpu-memory-utilization 0.96 (1) ·
64 GiB, GPU_UTIL 0.96 at PP8 (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)
power cap:
driver:
unlock build:
status:
tested by:
tested on:
source:
https://github.com/344303947/dsv41-flash-pp5-170hx (1) ·
https://github.com/arvkmr/vllm-170hx-glm5 (1) ·
https://github.com/iIIusi0n/qwen3.8-flash-next-cmp170hx (1) ·
https://github.com/kaka86mm/glm53-flash-170hx (1) ·
https://github.com/nguyenthimy2022kg-alt/Qwen-Flash-SM80-170HX (1) ·
https://github.com/zebgop-ops/qwen38-flashnext-pp (1)
Showing below up to 6 results in range #1 to #6.