The 3-step distill (speedup) LoRA is now baked into the diffusion model
itself, so the Moxie Multimedia Loader no longer applies any LoRA:
- moxie_loader.py: drop the models/loras scan, _auto_lora(), the module
level LoRA cache and _apply_lora(); the load chain is now
diffusion model -> attention backend -> CLIP -> VAEs. A leftover file
in models/loras is silently ignored and the folder can be deleted.
- moxie_project.py / moxie_preview.py: tooltips reference the baked-in
distill instead of the LoRA.
- tests + standalone loader script: assert no load_lora_for_models call
happens (the comfy.sd stub now raises if the loader tries), and fix
the pre-existing MMnode.tiny_vae stub gap (missing pack_vae_approx_files)
that broke the whole test module at fixture setup.
- README: models table drops the Turbo LoRA row; CHANGELOG entry added.
Node schema, outputs and sampling behavior are unchanged; workflows
saved before this change keep working.
Dropping ffmpeg.exe into python_embeded (next to python.exe) is a common
Windows support instruction; scan sys.executable's folder after the
portable root and CWD candidates.
Windows PATH edits only reach processes started afterwards, so a freshly
installed FFmpeg stays invisible to a running ComfyUI, and portable users
often drop ffmpeg.exe next to the run bat without any PATH entry at all.
get_ffmpeg_path now probes, after which() fails: the classic C:\ffmpeg\bin
and %LOCALAPPDATA% locations, winget Links / scoop shims / chocolatey bin,
the ComfyUI base folder and portable root (root, bin/, ffmpeg/bin/), and
finally the ffmpeg binary bundled with ComfyUI's own imageio-ffmpeg
dependency (ffmpeg only - imageio-ffmpeg bundles no ffprobe).
The H3 compose failure message now also points at concrete remedies
instead of only listing search locations.
Verified with scripts/test_merge_failure_reason.py: 19/19 stubbed checks
(portable-layout resolution, imageio-ffmpeg last resort, ffprobe exclusion,
and every compose failure reason).
merge_video_track_with_ffmpeg returned None for every failure mode (FFmpeg
missing, empty timeline, odd canvas, missing segment/locked-audio files,
spawn errors, non-zero exit), which collapsed into the useless
'FFmpeg could not compose the H3 project video' message users saw in the
ComfyUI error dialog.
Add an opt-in failure_reason collector: callers that need to surface the
cause (compose_h3_project_video) now pass a list and receive one specific
human-readable entry per failed precondition, or the FFmpeg stderr tail
(e.g. 'Unknown encoder libx264') when the command itself fails. The
nodes/basic.py tensor-merge fallback still uses the silent None contract
and is unchanged.
Verified with scripts/test_merge_failure_reason.py (16 stubbed checks:
every return-None path records a reason, compose propagates it).
The base-prefix computation produced '//' when the loader is served
from the top-level /extensions mount (the standard case), turning the
editor import into '//<pack>/index.js'. Browsers parse that as a
protocol-relative URL and resolve '<pack>' as a DNS hostname, so the
editor entry never loaded: console shows 'Failed to load resource:
net::ERR_NAME_NOT_RESOLVED moxie-multimedia/index.js' followed by
[vite:preloadError]. Only sub-path reverse-proxy deployments
accidentally worked. Base is now exactly '/' when there is no
sub-path prefix, and '<prefix>/' otherwise.
PyPI only hosts a stub sdist for nvidia-vfx whose wheel_stub build
backend aborts the entire 'pip install -r requirements.txt' run
(BackendUnavailable: Cannot import 'wheel_stub.buildapi'). The real
prebuilt wheels (win_amd64 cp310/cp311/cp312-abi3 + manylinux) live on
pypi.nvidia.com. Route nvidia-vfx there with file-level pip options
(--extra-index-url + --only-binary=nvidia-vfx) so pip always picks the
wheel and the stub is never built. Validated in a clean venv: the
wheel installs and provides the nvvfx module rtx_vsr.py imports.
HF shows 'empty or missing yaml metadata in repo card' when the
README has no YAML front matter. Add license (gpl-3.0) and discovery
tags at the top of README.md, matching the structure Comfy-Org uses
on its repackaged model repos.
The pack targets NVIDIA RTX 3060+ GPUs only, so speech (voxcpm),
subtitle recognition (openai-whisper, qwen-asr) and the RTX VSR
upscale (nvidia-vfx) move from on-demand extras into requirements.txt.
Verified on PyPI: all four resolve, torchcodec ships Windows wheels
for python 3.10-3.14, and the qwen-asr transformers==4.57.6 pin is
compatible with the rest of the set. requirements.txt is ordered
light -> heavy so re-running the install resumes cleanly.
The downloader script is removed; all model files live in the Hugging
Face version of the repository, so cloning it is the only step needed.
CHANGELOG also records the new requirements.txt from the previous
commit.
ComfyUI-Manager installs requirements.txt automatically, but the pack
never shipped one, so manual/portable installs got zero dependencies.
List the feature-level deps (soundfile, jieba, huggingface_hub,
ffmpeg-python, opencv-python, imageio) and document the embedded-python
install command for ComfyUI Portable plus optional extras (voxcpm,
openai-whisper, qwen-asr, nvidia-vfx).
- frontend/src + messages restored byte-identical to upstream v1.3.3:
format normalization shims, isMoxieEditorFormat gates, bundle marker,
tooltip reword, and their tests all removed; editor bundle rebuilt from
that pristine source so the editor behaves exactly like upstream
- format combo shows 'MiniMax' again (options=[MiniMax]); legacy queued
'Moxie Multimedia' values still normalized to MiniMax at execution
- T2V resolves natively again: t2v stays t2v in h3_task_type,
_multitrack_task_type and getMultiTrackTaskType; generation-mode t2v
mapping and the zero-media cloud fallback removed (upstream behavior);
Python tests and stub scripts updated accordingly
- editor delivery fix: core lets WEB_DIRECTORY overwrite the pack's
EXTENSION_WEB_DIRS entry, so the /extensions listing never contained
dist/release and a fresh browser never imported the editor (it only
ever appeared from stale browser cache). web/js/editor-loader.js now
imports the editor entry from the pack's own static mount, which
serves index.js with no-cache (chunks stay cacheable); the
/extensions/<pack>/index.js route was removed as a 404 trap
- shipped Moxie*.json templates carry format value MiniMax
Verified: frontend diff vs upstream = 0; bundle string checks + chunk
refs; vitest 565/566 (same pre-existing jsdom URL.revokeObjectURL env
failure); py_compile; test_t2v_r2v_and_loader.py ALL PASS;
test_rtx_upscale_integration.py ALL PASS
Root cause of the persistent missing Shot/Context combo on task segments 2+:
the 5e53555 rebuild ran before the rebase that brought the normalization
source in, so the shipped bundle still contained the pre-fix raw
format === 'MiniMax' gates (verified: zero 'Moxie Multimedia' literals in the
5e53555 chunks). The fix never reached users in that build.
- dist/release rebuilt from current source: normalization present, no raw
MiniMax gates in the editor chunk, marker 2026-09-28.1, chunk refs intact
- speakerReferenceTooltip no longer leaks the internal model name (en+zh)
- vitest 566/567 (same pre-existing URL.revokeObjectURL env failure)
ComfyUI core static-mounts every EXTENSION_WEB_DIRS entry at
/extensions/<name>/ late in server.add_routes(); our handler registers
earlier so aiohttp registration order gives it precedence. The editor
entry file is now revalidated on both plausible load paths.
- TaskSegmentEditor continuity (Shot/Context) combo, ref image size combo and
MiniMax prompt placeholder now use isMoxieEditorFormat: they stay available
even when the format widget value is missing or unrecognized (the pack ships
exactly one video format)
- __init__.py now serves dist/release/index.js with Cache-Control: no-cache:
the entry file keeps a stable URL while chunk names rotate every build, so a
stale cached entry file used to request deleted chunks (404) and silently
break the editor; hashed chunks remain cacheable
- bundle marker bumped to 2026-09-28.1; release rebuilt; vitest 566/567 (same
pre-existing URL.revokeObjectURL env failure)
- normalizeFormatValue now trims and accepts missing values; new
isMoxieEditorFormat helper treats the pack's single video format as
enabled even when the format widget value is empty or unrecognized
- TaskSegmentEditor normalizes its format prop once (continuity combo,
ref image size combo, MiniMax prompt placeholders, system prompt
template matching all follow); TrackArea audio lock and the
PreviewArea continuity/ref-size gates use the same helper
- the extension logs a bundle version marker on setup
([Moxie Multimedia Suite] editor bundle 2026-09-27.2) so a stale
browser-cached build is immediately visible
- release bundle rebuilt; vitest 566/567 (one pre-existing
URL.revokeObjectURL env failure on HEAD)
The dist/release bundle was last committed at repo init, so none of the
frontend changes since then were being served. The old bundle still
compared the raw format widget value against 'MiniMax', while workflows
now save the 'Moxie Multimedia' display label - which hid every
MiniMax-gated editor feature: the Shot/Context continuity combo, the
reference image size combo, the audio lock toggle, and MiniMax prompt
placeholders. Rebuilt with bun run build:release; bundle now contains
normalizeFormatValue at the widget read point and the T2V-to-R2V task
type resolver. vitest 566/567 (one pre-existing URL.revokeObjectURL env
failure on HEAD).
- T2V selections (editor default mode, explicit [start-end,t2v] override
tags, raw t2v values) now resolve to r2v in every task-type resolver
(h3_task_type, _multitrack_task_type, frontend getMultiTrackTaskType);
h3_generation_mode maps raw t2v to reference conditioning as belt and
braces. Reference-mode with zero reference media renders text-only
locally; the cloud path falls back to its native t2va request instead
of raising. i2v/l2v/v2v/edit flows unchanged.
- MoxieMultimediaLoader now presents a single model picker: LoRA, video
VAE, audio VAE, and CLIP widgets removed and auto-selected from the
pack's models/ folders (first sorted file, loud error when missing,
LoRA applied at strength 1.0 when bundled). Outputs and load order
unchanged; saved workflows keep working (extra widget values ignored).
Verified: py_compile; scripts/test_t2v_r2v_and_loader.py 50/50;
frontend vitest 566/567 (one pre-existing env failure on HEAD);
bun run build; scripts/test_rtx_upscale_integration.py all pass.
MoxieCombine now exposes exactly images / audio / fps - the leftover
VIDEO and FILENAME_PREFIX outputs are gone, so the output file name is
chosen on easy saveVideo itself. The base easy multitrackProjectVideoCombine
keeps its VIDEO and FILENAME_PREFIX outputs for legacy workflows.
The shipped Moxie-consolidated template is rewired accordingly: fps/images
links follow the new slot indices (2/0), the filename_prefix link is replaced
by the same project-scoped literal the removed output produced, and the
TEMPLATE's save location is unchanged.
pytest: new schema/execute/blocker coverage for MoxieCombine via a
_load_moxie_combine helper; base-combine assertions untouched.
- the format combo on MultiTrack Editor and MultiTrack Project now
shows 'Moxie Multimedia' (FORMAT_DISPLAY_LABEL); every queued value
is normalized back to the internal 'MiniMax' value at both execute
entries and check_lazy_status, so tracks_info, conditioning caches,
saved manifests and frontend lookups are unchanged
- frontend: normalizeFormatValue() maps the display label back at the
format-widget read point so MiniMax-specific editor features keep
working; shipped templates carry the new label
- removed the per-run 'Output upscale' console line; the restore now
happens silently (one-shot fallback diagnostics kept)
Note: workflows saved before this change need the format re-picked
once, since 'MiniMax' is no longer a combo option.
The compose step previously built the timeline at the manifest
resolution while staging segments are written at the (hidden) half-res
sampling size, so ffmpeg's overlay pasted the smaller segments into the
top-left quarter of a black full-res canvas and the combine's restore
gate saw frames already at the manifest size and never fired.
- compose_h3_project_video now probes the timeline segments and composes
at the size they all share (_uniform_probed_size; manifest dims remain
the fallback for mixed/unknown sizes)
- restore_project_video_resolution (utils/rtx_vsr.py) fast-probes the
composed file, and when it differs from the manifest decodes once,
upscales (RTX VSR, Lanczos fallback, exact size) and rebuilds the
video so VIDEO, images, audio and fps all carry the editor-selected
resolution; any failure keeps the original video
- the restore lives in the base combine so MoxieCombine (images slot)
and the legacy GetVideoComponents wiring are both covered; the
duplicated images gate in MoxieCombine is gone
Single-pass projects condition and sample at exactly half the
editor-selected resolution (plain integer halving, no dimension
policies). Moxie Project Video Combine restores the exact editor
resolution on the combined frames right before its images output,
driven by the fresh project manifest so a stale frontend widget can
never misgate it.
- utils/rtx_vsr.py: RTX VSR core (Apache 2.0, from
Comfy-Org/Nvidia_RTX_Nodes_ComfyUI) + Lanczos fallback; the gate is
simply 'frames differ from the manifest resolution'
- nodes/project.py: halve first-pass dims for single mode only;
dual-pass, SelfLift, passthrough and audio-only keep the editor size
- moxie_combine.py: restore before the images output
- scripts/test_rtx_upscale_integration.py: 17 standalone checks
Single-pass MultiTrack projects now condition and sample at half the
editor-selected resolution; Moxie Project Video Combine restores the
exact editor resolution on the combined frames before export, so the
editor resolution remains what the user sees and gets.
- utils/rtx_vsr.py (new): RTX Video Super Resolution core adapted from
Comfy-Org/Nvidia_RTX_Nodes_ComfyUI (Apache 2.0) driving the optional
nvidia-vfx package per frame, with automatic Lanczos fallback and
exact-size correction; owns the sampling-dimension policy and the
project_data gating so only half-res signatures are upscaled
- nodes/project.py: gate half-res conditioning/sampling to single-pass
mode (dual-pass, SelfLift, passthrough, audio-only unchanged)
- moxie_combine.py: restore editor resolution right before the images
output; VIDEO preview output passthrough unchanged
- scripts/test_rtx_upscale_integration.py: standalone sanity checks
(no ComfyUI/torch needed) for the dimension policy and gating
MultiTrack and H3 conditioning cache logging printed Chinese labels
(ProjectMedia/SegmentMedia statuses, timing and cache-size breakdowns).
Translate every console-facing string to English so the ComfyUI log is
fully English-readable:
- cache status values are now 'cache-hit' / 'first-load' (was Chinese),
including the _all_media_caches_hit sentinel comparison and the
_easy_media_cache_status data written into tracks_info
- cache size breakdown now reads 'file(...)=conditioning(...)+...'
with English category labels
- timing/stage messages, video-track compositing and locked-audio
extraction messages, and llama.cpp install errors are English
- fullwidth '|' in the media_loading timing line replaced with ASCII
- test assertions updated to the new English strings
No behavioral changes beyond the log/status wording; the Chinese
prompt-reference syntax (@图片N etc.) and LLM prompt templates are
intentionally untouched as they are functional.
Newer ComfyUI builds lock registered v3 node classes after registration
(comfy_api internal __setattr__ guard), so _apply_lora writing
cls._loaded_lora raised: Cannot modify class attribute '_loaded_lora'
on locked class 'MoxieMultimediaLoaderClone'.
The (path, sd, metadata) cache now lives at module level, which the
locking mechanism does not touch. Test monkeypatch updated to match.
The Moxie model set is MiniMax H3-only, so no other model choices are
exposed in the editor:
- Moxie Multimedia Loader: removed the clip_type dropdown (which listed
every CLIPType recipe in ComfyUI). The H3 Qwen3-VL text encoder is now
always loaded with CLIPType.MINIMAX; older ComfyUI builds raise a
clear error instead of silently loading the wrong encoder.
- Moxie Multimedia Loader: outputs are now exactly model, clip,
video_vae, audio_vae - the legacy FAST_MODEL_LOADER (model_loader)
bundle output is removed. The model output keeps the attention
backend and LoRA applied.
- Moxie MultiTrack Project: takes model, clip, video_vae and audio_vae
as required direct inputs instead of a bundled model_loader; the
fallback plumbing is removed while EasyMultiTrackProject keeps its
model_loader input for non-Moxie workflows.
- Moxie Preview Override: dropped the model_loader pass-through slots
and patches the model input directly.
- Moxie-consolidated.json reference workflow updated to the new slots.
- Added tests/test_moxie_nodes.py covering the schema contracts, the
attention -> LoRA -> output chain, the MINIMAX lock and the project
loader synthesis.
The upstream locales/zh/nodeDefs.json is not part of this repository,
so every test reading it failed at runtime. Guard all 16 call sites
with pytest.skip and add a tests/conftest.py that stubs folder_paths
so collection works outside a full ComfyUI installation.