feat(char-rp-gguf): max context — char-rp 96K, char-rp-reasoning 40K, q8_0 KV

Raise both RP seats to near-max context within the GPU0 budget using q8_0 KV cache
(near-lossless 8-bit, ~2x context/GB, flash-attn-backed). Verified coherent on both
(no Qwen KV-quant gibberish) at 64/50 tok/s.

- char-rp (Magidonia): 16K -> 96K (native 128K; 128K would starve the reasoning seat).
- char-rp-reasoning (QwQ): 16K -> 40960 (QwQ native max; beyond needs YaRN).
- kv_unified=true -> a single conversation gets the full n_ctx (slots share the pool).
- GPU0 ~93/97G, ~4.3G margin (gen fixed-util + static KV = stable, no OOM risk).
- New .env knobs: CHARRP_CTX / CHARRP_REASONING_CTX / CHARRP_KV_TYPE / CHARRP_REASONING_KV_TYPE.

stacks/char-rp-gguf/.env.example

@@ -24,7 +24,11 @@ CHARRP_PORT=8016
 CHARRP_MODEL=rp/TheDrummer_Magidonia-24B-v4.3-Q6_K.gguf
 # Alternate prose model (more literary flair, looser POV adherence) — uncomment to swap:
 # CHARRP_MODEL=rp/MS3.2-PaintedFantasy-v4.1-24B-Q6_K.gguf
-CHARRP_CTX=16384
+# Context. Magidonia native = 131072 (128K). 96K @ q8_0 KV ~ 8.4G VRAM; f16 KV doubles
+# that. Raise toward 131072 only if you drop the reasoning seat's ctx (shared GPU-0 budget).
+CHARRP_CTX=98304
+# KV cache dtype: q8_0 (~half VRAM, near-lossless) or f16 (max fidelity, 2x VRAM).
+CHARRP_KV_TYPE=q8_0
 
 # ── REASONING seat (char-rp-reasoning) ──────────────────────────────────────
 CHARRP_REASONING_CONTAINER=llama-charrp-reasoning
@@ -33,7 +37,10 @@ CHARRP_REASONING_PORT=8018
 # llama.cpp-managed CoT). ~50 tok/s @ Q5. Use Q6_K (~46 tok/s) for a touch more
 # quality if speed is not binding.
 CHARRP_REASONING_MODEL=rp/QwQ-32B-ArliAI-RpR-v4-Q5_K_M.gguf
-CHARRP_REASONING_CTX=16384
+# QwQ native ctx = 40960 (its max without YaRN). ~5.5G VRAM @ q8_0 KV.
+CHARRP_REASONING_CTX=40960
+# KV cache dtype (Qwen-arch): q8_0 verified coherent here; f16 if a future model gibbers.
+CHARRP_REASONING_KV_TYPE=q8_0
 # Thinking-token cap (QwQ over-thinks otherwise → starves the prose). 300-500 = a
 # concise, useful scene-plan before the response.
 CHARRP_REASONING_BUDGET=400

stacks/char-rp-gguf/compose.yaml

@@ -69,9 +69,15 @@ services:
       - --n-gpu-layers
       - "999"
       - --ctx-size
-      - "${CHARRP_CTX:-16384}"
+      - "${CHARRP_CTX:-98304}"
       - --flash-attn
       - on
+      # q8_0 KV cache ~halves KV VRAM (8-bit, near-lossless) → ~2x the context per GB.
+      # Mistral/Magistral handles q8 KV cleanly. Set f16 in .env to disable.
+      - --cache-type-k
+      - ${CHARRP_KV_TYPE:-q8_0}
+      - --cache-type-v
+      - ${CHARRP_KV_TYPE:-q8_0}
       - --jinja
     healthcheck:
       test: ["CMD-SHELL", "curl -fsS http://localhost:8080/health >/dev/null || exit 1"]
@@ -115,9 +121,16 @@ services:
       - --n-gpu-layers
       - "999"
       - --ctx-size
-      - "${CHARRP_REASONING_CTX:-16384}"
+      - "${CHARRP_REASONING_CTX:-40960}"
       - --flash-attn
       - on
+      # QwQ native ctx = 40960 (beyond needs YaRN → quality loss; don't). Qwen-arch can be
+      # KV-quant-sensitive: q8_0 is verified coherent here, but flip to f16 in .env if a
+      # future model shows gibberish.
+      - --cache-type-k
+      - ${CHARRP_REASONING_KV_TYPE:-q8_0}
+      - --cache-type-v
+      - ${CHARRP_REASONING_KV_TYPE:-q8_0}
       - --jinja
       # QwQ reasoning is template-native → llama.cpp manages it. --reasoning on surfaces
       # the trace in reasoning_content (content stays clean prose); --reasoning-budget