Frank 9971f83e8b perf: frank-default-small ctx 200000/stream, --no-mmap, batch 3072/1536 ...

Lower each stream to 200000 ctx (pool 600000) to free KV VRAM for the larger
batch. Add --no-mmap so the CPU-resident MoE experts load eagerly into RAM
(avoids decode page faults). Keep KV on GPU. Experts already fully on CPU.

2026-08-11 01:45:41 -07:00

..

deploy.sh

docs+deploy: Dankserver now llama.cpp frank-default-small (HauhauCS Q6_K_P + DFlash)

2026-08-11 01:18:29 -07:00

frank-default-large.service

feat: add Ling-3.0-flash vLLM service for Dankclaw (frank-default-large, port 7770)

2026-08-10 22:36:37 -07:00

frank-default-small.env

perf: frank-default-small ctx 200000/stream, --no-mmap, batch 3072/1536

2026-08-11 01:45:41 -07:00

frank-default-small.service

feat: frank-default-small = HauhauCS Qwen3.6-35B-A3B uncensored Q6_K_P + DFlash on dankserver

2026-08-11 00:57:04 -07:00

llama-server.service

sync scripts from remotes: ROCmFPX llama-server.service (dankclaw), dankserver vLLM serve/validate; drop failed-attempt scripts

2026-08-10 19:48:46 -07:00

serve-ling3-vllm.sh

feat: raise Ling-3 GPU memory utilization to 0.95

2026-08-10 23:25:02 -07:00

serve-llama-frank-small.sh

perf: frank-default-small ctx 200000/stream, --no-mmap, batch 3072/1536

2026-08-11 01:45:41 -07:00

serve-vllm-qwen36-dflash.sh

sync scripts from remotes: ROCmFPX llama-server.service (dankclaw), dankserver vLLM serve/validate; drop failed-attempt scripts

2026-08-10 19:48:46 -07:00

validate-vllm.sh

sync scripts from remotes: ROCmFPX llama-server.service (dankclaw), dankserver vLLM serve/validate; drop failed-attempt scripts

2026-08-10 19:48:46 -07:00