Drilldown: Recipe - CMP Wiki

Jump to content

Choose a table:

Recipe > page

: Qwen3.8 Flash Next on 3x 8 GB (clee PP3)

or Qwen3.8 Flash Next on 4x 8 GB

& cards: 8 GB

Use the filters below to narrow your results.

Qwen3.8-Flash-Next (2)

AWQ W4A16: wtdcode/Qwen3.8-Flash-Next-AWQ-W4A16 (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)

vLLM nightly 7d0b4e57 plus the author's pure-Python fork clee/vllm-sm80 (branch q38fn/stack) (1) · wtdcode/vllm-backport (image lazymio/vllm-backport:latest-sm80) (1)

latest-sm80 tag as of 2026-10-16 (1) · q38fn/stack on nightly-7d0b4e57 (1)

PP3, no P2P (works at PP2 on two cards) (1) · TP4 with expert parallel, MTP 3 (1)

1,000,000 (YaRN x4 from 262,144) (1) · 1,048,576 (YaRN x4) (1)

74.4 (1) · 116.49 (1)

2,101.4 (1) · 4,831 (1)

64 GB (1) · 64 GiB, gpu-memory-utilization 0.92 (1)

None (2)

None (2)

None (2)

Proven (1) · Reported (1)

c133 (1) · matstatman (1)

October 10, 2026 (1) · October 16, 2026 (1)

https://discord.com/channels/1518634379604136178/1528826330454429706/1560602001908170753 (1) · https://github.com/clee/qwen3.8-flash-next-3x170hx (1)

Showing below up to 2 results in range #1 to #2.

Q

Qwen3.8 Flash Next on 3x 8 GB (clee PP3)

Qwen3.8 Flash Next on 4x 8 GB

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