llama-swap: pin granite-4-small alongside qwen3.5-9b — stop swapping with qwen3.6-27b

Symptom: granite-4-small and qwen3.6-27b were evicting each other
when called in alternation. granite is the news-digest curator (fires
twice daily on cron) — being evicted means a cold reload (~5s) on
every digest tick, plus visible churn whenever the user uses 27b
concurrently.

Added granite-4-small to the `pinned` group as a persistent member.
~5-6 GB at Q4_K_M + 120K KV ≈ comfortable inside the existing pin
budget (qwen3.5-9b ~6 GB → ~12 GB total persistent). Single RTX 6000
Ada is 48 GB, leaves ~36 GB headroom for whichever non-pinned model
the user invokes (qwen3.6-27b at ~30 GB fits cleanly).

Updated the pinned group's docstring to capture the current member set
+ VRAM math + the historical context (qwen3.6-35-a3b was here, was
too heavy, got removed yesterday). Marked the granite ttl: 0 with the
matching "pinned — never unloads" comment as the other group members.

stacks/llama-swap/conf/config.yaml

@@ -419,7 +419,7 @@ models:
   "granite-4-small":
     name: "Granite 4.0 Small Q4_K_M"
     description: "IBM Granite 4.0 Small. Deterministic utility model for structured tasks."
-    ttl: 0
+    ttl: 0   # pinned — member of the `pinned` group, never unloads
     cmd: |
       /app/llama-server
       --context-shift
@@ -545,17 +545,27 @@ groups:
       - "qwen3-embedding-0.6B"
       - "qwen3-reranker-0.6B"
 
-  # Pinned general-purpose chat models. Coexist in VRAM, never unload.
-  # Members also have ttl: 0 individually so idle-timeout can't drop them.
-  # Currently just qwen3.5-9b — small enough (~6 GB at Q4 + KV) to be
-  # cheap to keep resident. qwen3.6-35-a3b was previously here but its
-  # ~29 GB at Q6_K_XL made other model loads OOM under memory pressure
-  # (persistent: true exempts members from eviction). It now sits with
-  # ttl: 0 outside any group — never idle-unloads, but the scheduler
-  # CAN evict it when another non-coexistent model is requested.
+  # Pinned general-purpose / utility models. Coexist in VRAM, never
+  # unload. Members also have ttl: 0 individually so idle-timeout can't
+  # drop them.
+  #
+  # Current pins:
+  #   qwen3.5-9b      — ~6 GB at Q4 + KV. General-purpose chat baseline.
+  #   granite-4-small — ~5-6 GB at Q4_K_M + 120K KV. Used by news-digest
+  #                     curator twice daily; pinning avoids the cold-load
+  #                     latency and prevents qwen3.6-27b (and similar)
+  #                     from evicting it when both are needed concurrently.
+  # VRAM budget: ~12 GB persistent in the pin slot. Single RTX 6000 Ada
+  # is 48 GB, so this leaves ~36 GB for whichever non-pinned model the
+  # user invokes alongside (qwen3.6-27b at ~30 GB fits cleanly).
+  #
+  # qwen3.6-35-a3b WAS in this group; removed 2026-04-27 because its
+  # ~29 GB at Q6_K_XL pushed concurrent loads OOM. Now lives outside
+  # with ttl: 0 — never idle-unloads but evictable under memory pressure.
   "pinned":
     swap: false
     exclusive: false
     persistent: true
     members:
       - "qwen3.5-9b"
+      - "granite-4-small"