Drilldown: Recipe

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) · Qwen3.8 Flash Next on 2x 8 GB (vektorprime PP2) (1)

GLM-5.3-Flash (320B MoE, 18B active) (2) · GLM-5.3-Flash (321B in the Maya README) (1) · Qwen3.8-Flash-Next (1)

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) · EXL3 3.05 bpw: turboderp/GLM-5.3-Flash-exl3; DFlash2 drafter converted to EXL3 6 bpw (1) · Maya-M GGUF (116 GB), every expert in VRAM; Maya-S (96.5 GB) and Maya-S24 (94.7 GB) also reported (1)

ExLlamaV3 (patched, pinned) (1) · ExLlamaV3 1.5.4 (built for SM80) with a small OpenAI-compatible server (1) · Project Maya (Strata and llama.cpp lineage, EXL3 loading, DFlash2 drafter) (1) · vLLM nightly + PR #53899 (a5530b9) + the repo's patch scripts (1)

None · recipe 0.2.1; engine 18305948ac · 0.1.dev1+ga5530b90c.d20260829 · 0.29.1rc1.dev402+ga5a30471f.ple1 · Deepinfer-SM80 2a828fc lineage; vLLM v0.28.1rc0 base · Deepinfer-SM80, vLLM v0.28.1rc0 base · Flun/glm53-flash-cmp170hx-exl3 at 5933112 · GNmac-lab at 2ad5741ab1; image ds41-cmp170hx:1.2.0-fidelity · latest-sm80 tag as of 2026-10-16 · not stated · q38fn/stack on nightly-7d0b4e57 · recipe 0.3.1; engine 18305948ac · recipe 1.0.0 (commit d4d1f525), engine a0c57b52 · recipe 1.0.0, engine a0c57b5237 · recipe 1.2.0 · recipe 1.8.0, engine 819146023f · recipe 31318fd; patches 0001-0020 · reports 0.0.0; patches/public-base-to-runtime.patch (109 files) · tag qwen38-recipe-1.0.0, engine 18305948ac · tag recipe-09 (309ef6d1); TabbyAPI deepseek-v41 (ce250ff) · upstream dev 0f685aa9fd plus 9 patches; repo db01c256 · v1.0.6 (this row) · vLLM 0.30.1rc1, fork commit 378c37b0 · vLLM v0.30.0 (ced6857a); plugin zankich/vllm v0.29.0z (6609bfef) · vllm-backport 3142ded1 · vllm-cmp170hx 18305948ac; TC-MoE fb9786ec · Other

8 GB (4)

2 (4)

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) · TP2 over NCCL (GPU P2P), DFlash2 K7 plus hybrid drafting (1)

None (1) · 262,144 (1) · 392,960 (profile q8_384k) (1) · 392,960 per request; 655,360-token Q8 KV pool shared by two requests (1)

60 (1) · 61.3 (1) · 75.4 (1) · 107.67 (1)

None (2) · 514 (1) · 1,533 (1)

64 GB (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 62.9 / 62.3 GiB idle, peaks about 64.4 / 63.0 GiB with vision (1)

None (3) · 180 W (author's cards); about 170 W with a modest memory OC in his Discord numbers (1)

None (3) · 610.57.04 (1)

None (3) · cmpunlocker (64 GiB); P2P-capable driver for the NCCL backend (1)

Proven (1) · Reported (3)

68x0 (1) · nestaa51 (1) · vektor_prime (1) · ZackO2o (1)

August 2026 (1) · October 2026 (2)

https://github.com/alexvnesta/glm53-flash-cmp170hx-tp (1) · https://github.com/Flun/glm53-flash-cmp170hx-exl3 (1) · https://github.com/mw00/project-maya#how-fast-is-it (1) · https://github.com/vektorprime/qwen38-flash-next-pp2 (1)