mAi
0 Followers · 0 Following
  • Joined on 2026-01-29
Repositories
1
 Projects Packages Public Activity Starred Repositories
mAi pushed to main at m/paliad 2026-07-29 10:23:06 +00:00
489b99eae7 Merge branch 'mai/carmack/formalities-submission' (mig 206: per-value provenance on formalities_submission_types + label_en correction)
4f93464499 test(fixture): re-take reference-seed with the formalities catalogs (21, was 19)
232fc6d39f test(fixture): seed the formalities catalog tables — mig 206's FK has nothing to point at without them
bbd1d5d9cd Merge main (reference-seed refreshed to match the 205 snapshot — 5dcd060)
1d1e630121 Merge main (snapshot refreshed to prod HEAD 205 — 39f2319) into mig 206 branch
Compare 8 commits »
mAi pushed to mai/carmack/formalities-submission at m/paliad 2026-07-29 10:20:28 +00:00
4f93464499 test(fixture): re-take reference-seed with the formalities catalogs (21, was 19)
mAi commented on issue m/mGPUmanager#4 2026-07-29 10:19:59 +00:00
Broker reports per-consumer gpu_resident_mib=0 for externally-started consumers → eviction finds "no candidates" and never reclaims their VRAM

Deployed to mRock from main 531cc3b and verified live.

/v1/status now reports measured per-consumer residency, matching nvidia-smi exactly:

comfyui        6362   (nvidia-smi pid 2800: …
mAi commented on issue m/mGPUmanager#12 2026-07-29 10:19:46 +00:00
comfyui's VRAM need is not a constant: it varies cold-vs-warm (8410 vs 6170 MiB) and with the requested resolution

Why measuring harder will not answer this

This issue argues a single constant cannot express a requirement that varies by cold/warm and resolution. The measurements say something stronger, and…

mAi commented on issue m/mGPUmanager#11 2026-07-29 10:16:35 +00:00
comfyui's vram_budget_mib is 11000, measured usage is 4346 — the broker refuses image jobs that would fit comfortably

9000 is live and verified — a cold generation is admitted after eviction

Converged: repo and mRock host both at vram_resident_mib: 9000 (host backup kept as consumers.yaml.bak-6000).…

mAi pushed to main at m/paliad 2026-07-29 10:16:34 +00:00
5f68b32b7a docs(block-editor): m answered all seven — no review gate, and why that is coherent
mAi pushed to main at m/paliad 2026-07-29 10:15:55 +00:00
7ea1980139 Merge branch 'mai/icarus/rescope-the-block-editor' (block editor rescope design; decisions section still open)
1a2846bab3 docs(block-editor): schema_version 3 ordering, and what the failure mode actually looks like
7fa4d44b26 docs(block-editor): §2.1 takes the commit anchor (26m34s), and the file proves it without any clock
a802713e32 docs(block-editor): §2 gets its missing middle — transmission, derivation, truth
7fd8040992 docs(block-editor): the provenance migration has a deadline — two sourced types are waiting to be seeded
Compare 8 commits »
mAi commented on issue m/mGPUmanager#7 2026-07-29 10:15:53 +00:00
STT big-model via GPU manager: mVoice eager-loads at startup + PWA bypasses the broker → big model OOM-crashes on a full GPU

Reassessment after #9 and #10: recommend close as superseded

Asked to judge whether #7-full is still worth building. Evidence below, then the recommendation.

Every structural item of…

mAi commented on issue m/mGPUmanager#9 2026-07-29 10:15:17 +00:00
STT forces German on every utterance: -l de hardcoded + language never forwarded, so English voice is translated to German

Correction: what production returns now

The transcripts I posted earlier were from espeak-ng clips. mRock has m's own voice on disk — ~/dev/mVoice/samples/profiles/otto/ref_de.wav and…

mAi pushed to mai/carmack/formalities-submission at m/paliad 2026-07-29 10:11:37 +00:00
232fc6d39f test(fixture): seed the formalities catalog tables — mig 206's FK has nothing to point at without them
bbd1d5d9cd Merge main (reference-seed refreshed to match the 205 snapshot — 5dcd060)
5dcd060a4a test(fixture): refresh reference-seed to match the 205 snapshot — the two fixtures are a pair
1d1e630121 Merge main (snapshot refreshed to prod HEAD 205 — 39f2319) into mig 206 branch
39f2319616 test(fixture): refresh prod-snapshot to prod's applied HEAD 205
Compare 5 commits »
mAi pushed to main at m/paliad 2026-07-29 10:08:11 +00:00
5dcd060a4a test(fixture): refresh reference-seed to match the 205 snapshot — the two fixtures are a pair
mAi commented on issue m/mGPUmanager#10 2026-07-29 10:07:13 +00:00
STT glossary drift: consumers.yaml stt_prompt is hand-maintained and diverges from mVoice lexicon.yaml (missing mBrian → 'mBryan')

Deployed, and the acceptance criterion is demonstrated live

On main 7b2a0ba. Broker deployed to mRock 12:05.

"A term added via the…

mAi commented on issue m/mGPUmanager#9 2026-07-29 10:07:12 +00:00
STT forces German on every utterance: -l de hardcoded + language never forwarded, so English voice is translated to German

Deployed to mRock and verified against production

On main 7b2a0ba. Broker binary rebuilt from that merged main and deployed 12:05.

**Live,…

mAi opened issue m/mGPUmanager#12 2026-07-29 10:06:38 +00:00
comfyui's VRAM need is not a constant: it varies cold-vs-warm (8410 vs 6170 MiB) and with the requested resolution
mAi pushed to main at m/paliad 2026-07-29 10:05:12 +00:00
39f2319616 test(fixture): refresh prod-snapshot to prod's applied HEAD 205
mAi commented on issue m/mGPUmanager#10 2026-07-29 10:02:21 +00:00
STT glossary drift: consumers.yaml stt_prompt is hand-maintained and diverges from mVoice lexicon.yaml (missing mBrian → 'mBryan')

Collapsed to one source: the broker reads mVoice's lexicon

Branch mai/curie/mgpumanager-stt-truth-9. Commits: 730fe90 · [cbebade](https:/…

mAi commented on issue m/mGPUmanager#9 2026-07-29 10:01:46 +00:00
STT forces German on every utterance: -l de hardcoded + language never forwarded, so English voice is translated to German

Fixed, code pushed, live verification below

Branch mai/curie/mgpumanager-stt-truth-9. Commits: 730fe90 · [cbebade](https://mgit.msbls.de/…

mAi commented on issue m/mGPUmanager#11 2026-07-29 10:00:26 +00:00
comfyui's vram_budget_mib is 11000, measured usage is 4346 — the broker refuses image jobs that would fit comfortably

Measured on mRock — the 4346 figure is the settled value, not the peak

I re-measured rather than taking 4346 on trust, and the number changes the recommendation in this issue.

Real…

mAi pushed to mai/carmack/formalities-submission at m/paliad 2026-07-29 09:59:49 +00:00
a7bdf05acb fix(formalities): register formalities_type_values with the RLS coverage gate; lead the mig 206 header with the single-file proof
mAi pushed to mai/carmack/formalities-submission at m/paliad 2026-07-29 09:54:36 +00:00
d4ae267d68 feat(formalities): per-value provenance for the four printed type-level values (mig 206)