Drilldown: Recipe - CMP Wiki

Jump to content

Choose a table:

Recipe > cards: 8 GB

& prefill tps: 1800-2500

Use the filters below to narrow your results.

DeepSeek V4.1 Flash heretic on 4x 8 GB (1) · DeepSeek V4.1 Flash on 4x 8 GB (GNmac) (1) · DeepSeek V4.1 Flash on 4x 8 GB plus RTX PRO 6000 (ExLlamaV3 TP5) (1) · GLM 5.3 on 8x 8 GB (vllm-170hx-glm5) (1) · Qwen3.8 Flash Next on 4x 8 GB (1)

DeepSeek-V4.1-Flash-1017 (2) · DeepSeek-V4.1-Flash-1017 heretic (pqhaz abliteration) (1) · GLM-5.3 (full 78-layer DSA MoE, not Flash) (1) · Qwen3.8-Flash-Next (1)

AWQ W4A16: wtdcode/Qwen3.8-Flash-Next-AWQ-W4A16 (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) · GLM-5.3-Int4-Int8Mix-AWQ-g64; drafter incoai/GLM-5.3-DFlash2 (1) · Hybrid: routed experts EXL3 (mostly 3 bpw, 2 bpw on layers 18-22) from LibertAI's 2.9 bpw pack; attention, shared experts, indexer and DSpark drafter official FP8; 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)

ExLlamaV3 fork (anoane/exllamav3-anemone) + TabbyAPI fork (anoane/tabbyAPI) (1) · Morrowmake/vllm-cmp170hx (vLLM fork, container image) (1) · Morrowmake/vllm-cmp170hx branch ampere-glm53 plus fail-closed source patches (1) · vLLM with a pinned sm_80 patch stack (built locally from source) (1) · wtdcode/vllm-backport (image lazymio/vllm-backport:latest-sm80) (1)

latest-sm80 tag as of 2026-10-16 (1) · recipe 1.0.0 (commit d4d1f525), engine a0c57b52 (1) · recipe 1.2.0 (1) · tag recipe-09 (309ef6d1); TabbyAPI deepseek-v41 (ce250ff) (1) · vLLM 0.30.1rc1, fork commit 378c37b0 (1)

PP8 (also PP9 and PP10), TP1 (1) · TP4 (1) · TP4 (default) or PP4 for x1/x4 links (1) · TP4 with expert parallel, MTP 3 (1) · TP5 across four CMP 170HX and one RTX PRO 6000 Blackwell (native backend, reduces through host memory); MTP draft on the PRO (1)

1,000,000 (YaRN x4 from 262,144) (1) · 1,048,576 FP16 (1) · 1,048,576 per request (1) · 262,144 (1) · 262,144 (up to 1,048,576) (1)

48.7 (1) · 99.9 (1) · 116.49 (1) · 158 (1) · 258 (1)

1,833 (1) · 2,101.4 (1) · 2,116 (1) · 2,416 (1) · 2,453 (1)

64 GB (1) · 64 GiB, --gpu-memory-utilization 0.92 (miner co-resident) (1) · 64 GiB, GPU_MEMORY_UTILIZATION 0.95 (1) · 64 GiB, GPU_UTIL 0.96 at PP8 (1) · gpu_split 58 GB per CMP, 70 GB on the PRO (1)

None (2) · 150 W per CMP (PRO 400 W) (1) · 170 W (author's 10-card run) (1) · 180 W (nvidia-smi -pl 180) (1)

None (3) · CUDA 13 capable; measured on 610.43.02 (1) · NVIDIA open 610.57.04 patched with Morrowmake/cmpunlocker 1f2c498 (profile 8gb) (1)

None (3) · cmpunlocker (64 GiB, 74 SMs) with a P2P-capable driver for TP4 (1) · Morrowmake/cmpunlocker 1f2c498; modprobe RmForceEnableGen2=1, RMPcieLinkSpeed=0x1, ForceP2P=0x11 (1)

Proven (2) · Reported (3)

anoane (1) · darkmaniac7 (1) · greencandle9753 (1) · matstatman (1) · reala7096 (1)

September 29, 2026 (1) · October 16, 2026 (1) · October 20, 2026 (1) · October 26, 2026 (1) · October 29, 2026 (1)

https://discord.com/channels/1518634379604136178/1528826330454429706/1560602001908170753 (1) · https://github.com/anoane/09-deepseek-v41-flash-1017-exl3-tp5-1m (1) · https://github.com/arvkmr/vllm-170hx-glm5 (1) · https://github.com/GNmac-lab/deepseek-v41-flash-1017-cmp170hx (1) · https://github.com/Morrowmake/dsv41-flash-cmp170hx-recipe (1)

Showing below up to 5 results in range #1 to #5.

D

DeepSeek V4.1 Flash on 4x 8 GB plus RTX PRO 6000 (ExLlamaV3 TP5)

G

GLM 5.3 on 8x 8 GB (vllm-170hx-glm5)

Q

Qwen3.8 Flash Next on 4x 8 GB

Retrieved from "https://cmp.wiki/wiki/Special:Drilldown/Recipe"