nodes/olivv-cs-ComfyUI-FunPack
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
531 Commits 3 Branches 11 Tags
2b8a81390d3d296ec908da170d820e5bb4ddcc7d
Commit Graph
10 Commits
Author SHA1 Message Date
digital-garbageandClaude Sonnet 5.5 8be3770080 feat: H3 tiled and X2 Detail VAE decode, phrase probe with Settings panel 
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
 2026-10-02 15:17:03 +03:00
digital-garbageandClaude Sonnet 5.5 f41136e3e0 fix: learners teach only from pushes that were felt, per setup; sla counter ignores throwaway calls; camera frame change; shot memory amount 0 
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
 2026-10-02 09:44:24 +03:00
digital-garbageandClaude Sonnet 5.5 4b9d92a5dd feat: late-branch guidance, STAS, camera move; shared rated_dial learner 
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
 2026-10-02 09:22:22 +03:00
digital-garbageandClaude Sonnet 5.5 7dab350961 feat: rating-driven steering rides into the next step's input (core/input_steer); block influence probe 
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
 2026-10-02 03:13:44 +03:00
digital-garbageandClaude Opus 5.5 3c187bda8a feat: first-step seed search; ALG passes the sampler call on instead of skipping inner wrappers 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
 2026-09-27 19:08:02 +03:00
digital-garbageandClaude Opus 5.5 895a61e8d3 feat: taste guidance + early taste guidance steer by a judge learned from ratings 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
 2026-09-27 18:41:02 +03:00
digital-garbageandClaude Opus 5.5 53cbc29948 feat: taste steering + taste attention learn from ratings; a failing wrapper no longer re-runs sampling 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
 2026-09-27 18:31:10 +03:00
digital-garbageandClaude Opus 5.5 3cd0c1e00a fix(v5): block repeat and shadow negative sample with the model compiler off 
Both replace what a block allocates, which crashes aimdo's compiler (seen on a
rental with v4's shadow negative). Attached to the model, so any sampler.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
 2026-09-27 17:59:18 +03:00
digital-garbageandClaude Opus 5.5 dc4d81be60 feat(v5): shadow negative module for H3 -- actually runs now 
v4 read the prompt length from transformer_options['c_crossattn'], which
ComfyUI never sets, so every block fell back to stock. It also stood down on
any i2v scene (per-token mod rows). Both fixed; tested on a real tiny H3.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
 2026-09-27 17:34:54 +03:00
digital-garbageandClaude Opus 5.5 c3ad0dfd79 feat(v5): randomizer and block repeat modules for H3, on shared chainable block hooks 
Block repeat's video-only mode now really keeps audio single-pass: H3 blocks
write in place, so v4 restored audio from an already-repeated stream.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
 2026-09-27 17:29:42 +03:00