frank-primary: shape G measured — arena 17.77 GiB / 7,762 slots, KV 401,408 tokens / 2.00x

Record the live numbers and the UTIL 1.0 verdict (unbootable: worker init
demands free > budget; ValueError on both ranks before a single weight loads).
Driver used 30.97 of 31.86 GiB; the 0.89 GiB/card margin held through a live
4,096-token generation at ~38-39 tok/s.
This commit is contained in:
Frank
2026-10-03 03:21:55 -07:00
parent c67a2aeb80
commit 47a2c1287b
8 changed files with 185 additions and 41 deletions
+7 -4
View File
@@ -2,10 +2,13 @@
#
# 2026-10-03 shape G (as shipped, after the 1.00 revert): the CUDA-graph plan
# term dropped from the analytic 0.55 GiB to the measured 0.07 GiB (0 base +
# 12 MiB/graph — the engine's own post-capture delta is -0.06 GiB), and UTIL at
# the 0.982 law. Arena 17.78 GiB / ~7,912 slots predicted (F measured 17.52 /
# 7,653). UTIL stays 0.982 ALWAYS: 1.00 is unbootable on this fork (worker init
# refuses budget >= device total) and 0.13-GiB-margin boots die under load.
# 12 MiB/graph), and UTIL at the 0.982 law. MEASURED on the live deploy: arena
# 17.77 GiB / 7,762 slots, KV 401,408 tokens / 2.00x concurrency (the first
# shape holding two full-length conversations with real spare), driver used
# 30.97 GiB of 31.86 — 0.89 GiB of real free VRAM per card, held through a
# live 4,096-token generation. UTIL stays 0.982 ALWAYS: 1.00 is unbootable on
# this fork (worker init refuses budget >= device total) and 0.13-GiB-margin
# boots die under load.
# `budget - non-expert - KV reserve - graphs - overhead` and the KV reserve is a
# registry constant, so those two terms are the whole flag surface.
# 2026-10-03 shape F (absorbed): MNBT 2048 + the graph term pinned. Measured on