Drilldown: Recipe
More actions
Recipe > decode tps:
90-100
& source
:
https://github.com/DUFRelic/qwen38-flash-next-recipe
or
Other 
& source
:
https://github.com/DUFRelic/qwen38-flash-next-recipe
or
Other 
Use the filters below to narrow your results.
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) ·
Native MXFP4 experts + FP8 attention (official deepseek-ai checkpoint, about 155.4 GiB); KV cache fp8 (2)
ExLlamaV3 fork (anoane/exllamav3-anemone) + TabbyAPI fork (anoane/tabbyAPI) (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)
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) ·
tag recipe-09 (309ef6d1); TabbyAPI deepseek-v41 (ce250ff) (1)
PP3 (VLLM_PP_LAYER_PARTITION=15,15,13) with DSpark, 5 draft tokens; PCIe Gen2 x4, no P2P (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) ·
TP5 across four CMP 170HX and one RTX PRO 6000 Blackwell (native backend, reduces through host memory); MTP draft on the PRO (1)
Showing below up to 4 results in range #1 to #4.

