digital-garbageandClaude Sonnet 5.5 5103ec65fa Test: H3 LoRA files (Diffusers and native layouts) land exactly on H3-Image ...

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

2026-11-11 03:51:29 +02:00

..

conditioning

Alternating prompts and Step merging: the Inactive verdict covers both stages of a Two stages run and comes after the last

2026-11-10 21:01:23 +02:00

latent

Review: only H3 LoRAs are renamed to H3 names (LTX transformer_blocks LoRAs load again)

2026-11-05 15:13:03 +02:00

loaders

Reference sheet generator (optional module + dialog); leak report tests start from unwrapped ComfyUI functions

2026-11-10 00:50:31 +02:00

media

Reference sheet generator (optional module + dialog); leak report tests start from unwrapped ComfyUI functions

2026-11-10 00:50:31 +02:00

models

Test: H3 LoRA files (Diffusers and native layouts) land exactly on H3-Image

2026-11-11 03:51:29 +02:00

output

Plain short hints for the model loader and other long settings; details moved to the docs; a test fails on hints over 200 characters

2026-11-10 00:38:26 +02:00

sampling

MiniMax H3-Image: REINS and the Attention explorer work on it; the explorer skips the sound and clip regions on a model without sound

2026-11-11 03:31:22 +02:00

system

Rating: a clip the Shot camera learned from is no longer reported as untaught

2026-11-02 13:09:07 +00:00

training

Reference sheet generator (optional module + dialog); leak report tests start from unwrapped ComfyUI functions

2026-11-10 00:50:31 +02:00

__init__.py

core+shell: the bridge, and the first module through it

2026-08-28 19:58:06 +03:00

_core.py

Sparse attention method veda for MiniMax H3 (experimental), with an optional predictor download

2026-11-05 21:32:39 +02:00

.gitkeep

core: serve the app, with the module allowlist as the wall

2026-08-28 14:18:55 +03:00