docs: frank-default-small — 4 parallel slots, batch 8192/2048, n-cpu-moe 32

Update Dankserver System Specs and deployment to reflect the 4-stream config:
- 4 parallel slots (800000 total ctx), batch 8192 / ubatch 2048
- n-cpu-moe 32 (8 expert layers back on GPU)
- ~870 MiB VRAM headroom under 4-stream load
- Knob documented: n-cpu-moe 36 gives ~1.7 GiB more headroom if needed

docs/Dankserver System Specs.md

@@ -51,11 +51,11 @@ Dankserver is an NVIDIA GPU server for vLLM inference. The host runs Ubuntu 24.0
 
 ### Active Services
 
-The active `frank-default-small` service (llama.cpp, `b10108+`) serves the HauhauCS Qwen3.6-35B-A3B uncensored `Q6_K_P` model on port 7770 with **ngram-mod** speculative decoding. It runs 3 parallel slots at 200000 context each (pool 600000). All 40 layers' MoE expert weights sit on CPU (`--n-cpu-moe 40`) while non-expert tensors and the full-attention KV cache stay on GPU (`--gpu-layers all`, KV K=`q8_0`, V=`f16`). It uses batch 3072 / ubatch 1536 with `--ctx-checkpoints` and `--no-mmap`. At load it uses about 14 GB of the 24 GB VRAM and about 32 GB of RAM.
+The active `frank-default-small` service (llama.cpp, `b10108+`) serves the HauhauCS Qwen3.6-35B-A3B uncensored `Q6_K_P` model on port 7770 with **ngram-mod** speculative decoding. It runs **4 parallel slots** at 200000 context each (pool 800000). Of the 40 layers' MoE expert weights, **32 sit on CPU** (`--n-cpu-moe 32`) while the remaining **8 layers' expert weights** are resident on GPU along with all non-expert tensors (`--gpu-layers all`). KV stays on GPU with K=`q8_0`, V=`f16`. Batch 8192 / ubatch 2048, `--ctx-checkpoints 4`, `--no-mmap`. At idle ~23.5 GB VRAM used; under a 4-stream 300-token burst peaks ~23.7 GB, leaving **~870 MiB** headroom. Idle RAM ~32 GB of 62 GB.
 
 ## Hardware Relevance
 
-- 24 GB VRAM limits the GPU-resident footprint. With 3 x 200000 context plus batch 3072/1536 activation buffers, the non-expert weights and KV cache consume about 14 GB, leaving about 10 GB of headroom. MoE experts stay on CPU; `--no-mmap` preloads expert tensors into RAM, eliminating decode page faults (uses ~32 GB of the 62 GB available).
+- 24 GB VRAM is almost fully saturated with this config. At idle ~23.5 GB used; under a 4-stream 300-token burst peaks at ~23.7 GB, leaving **~870 MiB** headroom. The 8 layers' expert weights back on GPU account for the tight margin. If headroom proves too thin for production bursts, dial back `--n-cpu-moe` to 36 (only 4 layers on GPU, ~1.7 GiB more headroom) by editing `scripts/frank-default-small.env`.
 - 62 GB system RAM holds the ~29 GB of MoE expert weights with room to spare (~57 GB available after load).
 - 127 GB swap is available for CPU KV cache paging if GPU KV offload is disabled.
 - Root disk has approximately 140 GB free. Store the GGUF models under `/home/dank/models` (about 32 GB total).
@@ -65,4 +65,4 @@ The active `frank-default-small` service (llama.cpp, `b10108+`) serves the Hauha
 - [[VLLM Overview]]
 - [[Hybrid CPU GPU Processing Dankserver]]
 - [[RAM CPU Offloading in VLLM]]
-- [[Running Qwen3.6 35B A3B FP8 on Dankserver]]
+- [[Running Qwen3.6 35B A3B FP8 on Dankserver]]