docs(scriberr): slicer patch live on fv-ml1 as local-blackwell-a353078-slicer1; live GPU 1 peak 5,496 MiB

Deployed 2026-09-30 1211 PT by pointing SCRIBERR_IMAGE at the patched tag (.env backed up as .env.bak-20260930-pre-slicer1; rollback is the unpatched scriberr:local-blackwell). PrepareEnvironment rewrote the env's parakeet_transcribe_buffered.py from the embed (sha256 matches the patched source). One live run on GPU 1 beside intern-decision peaked at 5,496 MiB. Memory records the open Parakeet mid-chunk dropout finding and the held upstream PR.
This commit is contained in:
vh
2026-09-30 12:14:09 -07:00
parent ee3db68db1
commit 3c5f1ea803
3 changed files with 10 additions and 6 deletions
@@ -191,6 +191,7 @@ opt-in `--pause-search`, off by default.
| shipped script (v3, final), 120 s, p1 | 3 | 5,496 / 5,496 / 5,496 MiB; all 3 CLI outputs byte-identical to the harness |
| shipped script, 120 s, scotus (public) | 3 | 5,496 / 5,496 / 5,496 MiB |
| positive control: shipped script, **300 s**, scotus | 1 | 7,056 MiB (deterministic; 300 s was n=3 in the earlier table) |
| **live, GPU 1**, deployed container, `docker exec` as appuser, 120 s, p1 | 1 | **5,496 MiB** (2026-09-30 1212 PT, beside intern-decision; 49 s for 35 min of audio; seam check OK) |
Zero spread; the peak is set by the 120 s maximum, and the overlap sits inside
it. The scotus file reads the same peak as p1, so the rebuild script uses it