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.
Bigger logical/physical batches for prefill throughput. Enable
--ctx-checkpoints to park attention checkpoints in RAM, holding peak GPU
memory down so the larger ubatch fits the 24GB VRAM alongside 3x262144 KV
without OOM. Weight split unchanged (experts CPU, non-experts GPU).
llama.cpp splits --ctx-size across --parallel slots, so 262144 total gave
87552 per stream. Non-negotiable is 262144 per stream: total pool must be
786432.
llama.cpp service replacing retired vllm-awq-qwen36 on port 7770 with alias
frank-default-small. Wrapper sources frank-default-small.env for tunables:
3 parallel slots at 262144 ctx, draft-dflash n-max 7, KV K=q8_0 V=f16 on GPU,
all MoE experts on CPU (--n-cpu-moe 40), non-expert tensors on GPU, mmproj
vision + jinja. Unit: frank-default-small.service (ExecStart in dankserver
repo clone /home/dank/vLLM/scripts).
Frank