r35 promoted to GLM 4x production (operator decision); r31 stays as rollback lane

history/night-shifts/2026-09-12.md

@@ -95,6 +95,16 @@ exact capture, SEQS 8, thinking+max, engram-local, deepseek-v4-flash alias).
 - Gist (published daytime): https://gist.github.com/kilork/1ba224bfe28779c571d096cf954fff31
   — base recipe card + live b12x overlay instructions in the README.
 
+## Production state after the daytime window
+
+- **Production: `glm53-flash-r35-4x-dcp4`** (4nodes-switch, promoted 09-12
+  per operator decision — r35 stays; r31 recipe remains as the rollback
+  lane, image tagged on all nodes).
+- Follow-ups added: (a) r31 re-bench on dev65 at the next 4x window for a
+  harness-clean delta (and to settle TC-51 variance-vs-regression);
+  (b) 32K-prefill probe if the split-MoE prefill gain on GB10 ever matters
+  (outside the current bench matrix).
+
 ## Morning follow-ups (ranked)
 
 1. Aiden `production-1.2` on Hub (DCP2-1.2 with the launcher env values) —

specs/tech-architecture/glm53-recipe-state.md

@@ -1201,5 +1201,7 @@ recompile on first boot):
   fail (step-ordering), TC-35/TC-40 partials. Perf flat (their +8.8% prefill
   gain was a 32K-prefill RTX cell — outside our matrix's 8192 max prefix).
   KV 8,060,928 tokens @512K (15.4x concurrency). TC-51 = variance vs
-  regression unresolved. Full log: `history/night-shifts/2026-09-12.md`
-  ("Daytime follow-up: R35 adoption + A/B").
+  regression unresolved. **Promoted to production 09-12** (operator: r35
+  stays; r31 = rollback lane, image tagged on all nodes). Full log:
+  `history/night-shifts/2026-09-12.md` ("Daytime follow-up: R35 adoption
+  + A/B").