Drilldown: Recipe
More actions
Recipe > decode tps:
20-80 
& source :
https://github.com/alexvnesta/glm53-flash-cmp170hx-tp or
Other
& source :
https://github.com/alexvnesta/glm53-flash-cmp170hx-tp or
Other Use the filters below to narrow your results.
page:
DeepSeek V4.1 Flash on 2x 8 GB (Project Maya) (1) ·
DeepSeek V4.1 Flash on 6x 8 GB (TP6) (1) ·
GLM 5.3 Flash on 2x 8 GB (ExLlamaV3 layer split) (1) ·
GLM 5.3 Flash on 2x 8 GB (Project Maya) (1) ·
GLM 5.3 on 8x 8 GB (Int4-Int8Mix PP8) (1) ·
GLM 5.3 on 8x 8 GB (vllm-170hx-glm5) (1) ·
Qwen3.8 Flash Next on 2x 8 GB (vektorprime PP2) (1) ·
Qwen3.8 Flash Next on 3x 8 GB (clee PP3) (1)
model:
quant:
AWQ INT4: cyankiwi/Qwen3.8-Flash-Next-AWQ-INT4; 95 GiB BF16 PLE table in host RAM (FP8 table optional) (1) ·
EXL3 3.05 bpw: turboderp/GLM-5.3-Flash-exl3 (1) ·
GLM-5.3-Int4-Int8Mix (INT4 / INT8 mix); KV fp8_ds_mla (1) ·
GLM-5.3-Int4-Int8Mix-AWQ-g64; drafter incoai/GLM-5.3-DFlash2 (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) ·
Official checkpoint: dense MXFP8 through an SM80 Marlin W8A16 path, routed experts MXFP4 through W4A16; KV 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)
engine:
Custom SM80 vLLM (patched from public commit 79a7108d9) (1) ·
ExLlamaV3 1.5.4 (built for SM80) with a small OpenAI-compatible server (1) ·
Morrowmake/vllm-cmp170hx branch ampere-glm53 plus fail-closed source patches (1) ·
Project Maya (Strata and llama.cpp lineage) (1) ·
Project Maya (Strata and llama.cpp lineage, EXL3 loading, DFlash2 drafter) (1) ·
vLLM (build not stated) (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)
engine version:
cards:
topology:
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 with PLE CPU offload (1) ·
PP3, no P2P (works at PP2 on two cards) (1) ·
PP8 (also PP9 and PP10), TP1 (1) ·
PP8, TP1 (1) ·
TP6 (virtual-head padding), PP1 (1)
context:
decode tps:
vram per card:
64 GB (2) ·
64 GB, --gpu-memory-utilization 0.95 (1) ·
64 GiB, --gpu-memory-utilization 0.97 (1) ·
64 GiB, gpu-memory-utilization 0.92 (1) ·
64 GiB, GPU_UTIL 0.96 at PP8 (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)
power cap:
driver:
unlock build:
status:
tested by:
tested on:
source: (Click arrow to add another value)
Showing below up to 8 results in range #1 to #8.