Frank 6a7025d01e Iteration 2: Remove --enforce-eager to enable torch.compile (+14.6% prompt speed: 5.95 -> 6.82 tok/s)

2026-08-07 23:07:51 -07:00

validate-vllm.sh

Initial vLLM Qwen3.6-35B-A3B-FP8 serving with CPU expert offloading

2026-08-07 22:20:52 -07:00

vllm-qwen36.sh

Iteration 2: Remove --enforce-eager to enable torch.compile (+14.6% prompt speed: 5.95 -> 6.82 tok/s)

2026-08-07 23:07:51 -07:00

S

Description

Podman deployment configs for Dankserver and Dankclaw

8.2 MiB

Languages

Shell 60.4%

Python 21%

HTML 6.2%

Jinja 3.9%

C++ 3.9%

Other 4.6%