Hanashi and Claude Fable 5
e375345bf6
SERVING-PLAN: per-output-second cost and pricing at all margins
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 20:08:03 -04:00
Hanashi
521468fc36
gitignore
2026-08-06 20:06:03 -04:00
Hanashi and Claude Fable 5
111f748c9d
SERVING-PLAN.md: paid/free pools, pricing at 0-70% margins, capacity scenarios
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 20:03:21 -04:00
Hanashi and Claude Fable 5
77524a10e6
final: 5090 tables (incl. bf16-streaming), stats tooling covers all 5 datasets
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 19:21:16 -04:00
Hanashi and Claude Fable 5
056d5785e7
MITIGATIONS.md: 24GB multi-ref RESCUED — full envelope verdicts per machine class
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 19:15:36 -04:00
Hanashi and Claude Fable 5
e33192a283
installer: stage bench images on the volume — immune to fetch rate limits
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 18:02:49 -04:00
Hanashi and Claude Fable 5
9fc70a0b81
MITIGATIONS.md: per-machine-class OOM map — RAM cap decides mitigation burden
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 17:09:40 -04:00
Hanashi and Claude Fable 5
a88626c0ca
bench: refmatrix suite — ref count (1-9) x ref resolution ladder
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 16:34:35 -04:00
Hanashi and Claude Fable 5
a4bd386e52
BENCHMARKS.md: final cross-GPU tables, failure stats, multi-ref rescue status
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 15:23:36 -04:00
Hanashi and Claude Fable 5
c0a875423e
start-comfyui: expandable_segments allocator config (fragmentation OOM hygiene)
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 15:11:18 -04:00
Hanashi and Claude Fable 5
4c2635515b
bench: --ref-downscale option — pre-shrink refs before the node (24GB OOM mitigation)
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 14:44:54 -04:00
Hanashi and Claude Fable 5
a293d0a2c4
reliability: failure-stats + output-validation tooling; persistent output dir
...
- bench/failures.py: per-GPU success rates and error-class breakdown by config
- bench/validate_outputs.py: ffprobe every recorded output via /view
- finding: 24GB cards fail on multi-ref (TE vision OOM), not duration/size;
100% of successful runs produce valid videos
- outputs now written to /workspace (survives pod resets)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 14:41:00 -04:00
Hanashi and Claude Fable 5
fbc68bb749
concurrency experiment measured: co-scheduling 13% slower than sequential — one worker per GPU confirmed
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 14:27:32 -04:00
Hanashi and Claude Fable 5
24134b9653
BENCHMARKS.md: B200-complete results — bf16 fastest family, ref scaling, cross-GPU interim
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 14:16:13 -04:00
Hanashi and Claude Fable 5
e90f64b30e
bench: results summarizer (mean±std tables); rebuild CSVs from authoritative JSONL
...
The CSV header predated the ref_count column so later rows were misaligned;
JSONL rows were always complete.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 13:51:17 -04:00
Hanashi and Claude Fable 5
53b4a133e2
bench: survive server loss during polling (mark run failed, keep suite alive)
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 03:48:37 -04:00
Hanashi and Claude Fable 5
427f2878c4
start-comfyui: per-host logs (shared volumes!), COMFY_ARGS passthrough
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 03:31:45 -04:00
Hanashi and Claude Fable 5
58b4755475
start-comfyui: wait out mid-shutdown process before already-running check
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 03:23:09 -04:00
Hanashi and Claude Fable 5
64e488fa49
start-comfyui: nohup fallback when tmux unavailable; parameterize port-forward
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 03:04:28 -04:00
Hanashi and Claude Fable 5
3082ee796f
installer: fetch real bench images during pod bootstrap
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 03:01:30 -04:00
Hanashi and Claude Fable 5
1debde7673
bench: v2 reference set — JP stills, Ashitaka/Bebop, forests, volcanos (up to 23k px)
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 02:45:34 -04:00
Hanashi and Claude Fable 5
94a94a268f
docs: concurrency & cost-per-video report across GPU tiers
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 02:28:24 -04:00
Hanashi and Claude Fable 5
c3cda1fb30
WEIGHTS.md: all 9 files persistent on 1TB volume, tmpfs scheme retired
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 02:12:42 -04:00
Hanashi and Claude Fable 5
cc111c0c33
bench: real reference images (anime/film/panorama) replace synthetic gradients
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 02:10:27 -04:00
Hanashi and Claude Fable 5
4a802d8c3e
bench: key-config suite with N=3 repeats for statistical averages
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-08-06 02:07:28 -04:00