docs: add recommended V3.5 workflow template
This commit is contained in:
@@ -7,6 +7,7 @@
|
||||
- Added experimental `H3 Continuum Hi-Res Fix V3.5`, using a bounded pixel/VAE resize round trip before low-denoise Second Pass sampling.
|
||||
- Added `H3 Continuum Latent Resize V3.5` as an advanced utility; direct high-ratio latent interpolation is not the recommended main Hi-Res Fix path.
|
||||
- Added `H3 Continuum Assemble + Seam V3.5` with Auto, RAM, and Windows-safe disk-backed video buffers while keeping audio in RAM.
|
||||
- Added the recommended V3.5 template workflow with optional Hi-Res Fix, Auto assembly, shared still-image megapixel sizing, and separate VHS Video Reference sizing.
|
||||
- Preserved Exact Duration, seam behavior, physical decode-group ordering, first-pass audio passthrough, and external Core VAE Decode contracts.
|
||||
- Kept V3.4 nodes and saved-workflow compatibility unchanged.
|
||||
- Validated a 9.49 GiB final IMAGE with hash-identical RAM/Disk-backed output and approximately 9.50 GiB lower Windows private memory in Disk-backed mode.
|
||||
|
||||
+4
-3
@@ -2,7 +2,7 @@ dca65fb1393e67f59c6f81f9411e3e745eb8f1097e9a2adf7ffc6f4e3901ca9f __init__.py
|
||||
15b900d02624bea6cfe8ea25b22266a0c68b38b9f10c9089be0e38988bf2e34d .github/workflows/tests.yml
|
||||
0a2f69d4cec89d440f5d5aa1ec68a7c9a741382d313f80dcddf96d13f70b3e68 .gitignore
|
||||
16ea21aa8fe115ce8f6538a0268bc64762be2fd6931b3a0071a86a8cc9d218cf AGENTS.md
|
||||
7b5c054c57940c7828c7ae5712cf796211b6a3bee0c50cc47fac03c3fffa0356 CHANGELOG.md
|
||||
2c370cc502c9badf4a9d8c6663af28643017a85d6a115d6d3e052725afb2b879 CHANGELOG.md
|
||||
e97e47c7585e81d9ec5fd32c0ca0b64699fd4610cd03e9d6575a91c30a10ca4e compatibility.py
|
||||
4d62508f900af9f93ec320afb900c27cc9a300d1e0c0a94714258498c5ede5fc conditioning.py
|
||||
71e3df269e226b1ecad597f9831c2a10b2215c8a3783652e9e9f536e99ce46c9 constants.py
|
||||
@@ -52,6 +52,7 @@ b8275c8cecde12aee0418cea456d1e7f4b875cbb1bce32f1f3d9e06a44dd0be1 examples/v2_pr
|
||||
4e410aed8d7f6ebf11c98d4f71571e46855e4619c2bcbec921df70f29189b6b1 examples/workflows/MiniMax_H3_Continuum_V33.json
|
||||
5eab33c110e07ec08ac66f289f20c295a90ca026ea5231ad6121755a28a2ded8 examples/workflows/MiniMax_H3_Continuum_V34_turbo.json
|
||||
660e955c983e41947f5abfd89e66fbe5759ffda01e6c1a7164d580b0c41b2120 examples/workflows/MiniMax_H3_Continuum_V34.json
|
||||
ed31ddd3d0af5e6f8dbd609e5ef51872322dbe52eaad8cae35e16342ac53dbfa examples/workflows/MiniMax_H3_Continuum_V35.json
|
||||
e8f41dee2240f1fcc6e8fd12448856513483ff3ae0427b2c554bb351f9ae8047 graph_contract.py
|
||||
3d14acac341c63207995e2bd45cd407fdb3e2308273b1c90605adfe6f2e0bd8f hardening.py
|
||||
6abdbc29472c7f9d447085b189fc47b69e110b94ef83192f9689d04000408843 INSTALL_JA.txt
|
||||
@@ -68,8 +69,8 @@ a7084bc1d12ab25a1ebe29a089dba72d6c1147e0f5556837e6321277562ad1f5 PACKAGE_VALIDA
|
||||
b470819f16fc179654dc1600721a63f0a5a14e777f4c004d574b499da6886ba5 public_api.py
|
||||
16bc48b9ef6127af7e0e3ddd96791640cedf5159cd7c06452d72604f41998a52 pyproject.toml
|
||||
94fdc2d545b2d4077a4d70229fd22ba3ccac21d49d1097d2873f76d04e9a3c7a pytest.ini
|
||||
365f1b1ecc89f8f2ced338ce467d02384ef612441202a15064e5d1a5c6cd1bcf README_JA.md
|
||||
18d3e7302db1bfe4f4417f6c520d2fcc616dceb23f703eb9e733147276b9c9a2 README.md
|
||||
02d76a521f02c5d1396ce81a6c1f889c690ace97647998679c6ce9de84fc56ae README_JA.md
|
||||
2e45803bbd1cdeb32656b71c1f9e78c7da0563bf1b1a3d5e2fd27bb734159347 README.md
|
||||
5d89c366b4f076fa73a2695557e0af749bf95df1dad0d55dbb2fe5b1c2519ad7 reference_audio.py
|
||||
8f1744dc99f1a3a222b7d6e986f6dac1edcc0abbbe8b61a1d2606b1ca3569b3d reference_video.py
|
||||
2995924fcb7bac27b548b01187d8276d2ab103c6f43416cf85400de3846c6297 reference.py
|
||||
|
||||
@@ -12,6 +12,14 @@
|
||||
|
||||
Updated: 2026-08-24
|
||||
|
||||
## V3.5 recommended template workflow (2026-08-24)
|
||||
|
||||
- `examples/workflows/MiniMax_H3_Continuum_V35.json` is the recommended V3.5 starting workflow.
|
||||
- Hi-Res Fix is disabled by default and `H3 Continuum Assemble + Seam V3.5` uses Auto backend selection.
|
||||
- First Frame, Last Frame, and Reference Images 1-3 share one 0.3 MP control. Video Reference remains independently sized by the VHS loader.
|
||||
- Reference Images 1-3 and Video Reference are routed to their matching V3.5 sampler sockets; `refine_context` and `video_vae` are connected to the integrated Hi-Res Fix path.
|
||||
- V3.4 standard and Turbo templates remain available for saved-workflow compatibility.
|
||||
|
||||
## Baseline
|
||||
|
||||
- Current public baseline: V3.4.0
|
||||
|
||||
@@ -122,10 +122,13 @@ This is a usability and reliability decision, not a claim that the experimental
|
||||
|
||||
## Example workflows
|
||||
|
||||
- [V3.5 recommended template](examples/workflows/MiniMax_H3_Continuum_V35.json)
|
||||
- [V3.4 standard](examples/workflows/MiniMax_H3_Continuum_V34.json)
|
||||
- [V3.4 Turbo](examples/workflows/MiniMax_H3_Continuum_V34_turbo.json)
|
||||
|
||||
The V3.4 templates remain valid in V3.5. A release-grade V3.5 UI workflow JSON is not bundled yet; use the connection guide above rather than older diagnostic workflows.
|
||||
The V3.5 template is the recommended starting point. Hi-Res Fix is disabled by default, the V3.5 Assemble backend is set to Auto, and First Frame, Last Frame, and Reference Images 1-3 share one megapixel control. Video Reference sizing remains controlled by the VHS loader. Advanced users can bypass or rearrange the optional sections as needed.
|
||||
|
||||
The V3.4 templates remain valid for saved-workflow compatibility.
|
||||
|
||||
The templates include optional external nodes such as Spectrum, Video Helper Suite, rgthree, EasyUse, and RTX Video Super Resolution. Install, replace, connect, or bypass them according to your installation. Media and acceleration choices remain under user control.
|
||||
|
||||
@@ -538,4 +541,4 @@ No OOM was observed in the cited recent local V3.4 checks, including two-chunk 8
|
||||
|
||||
## License
|
||||
|
||||
See [LICENSE](LICENSE).
|
||||
See [LICENSE](LICENSE).
|
||||
|
||||
@@ -70,6 +70,12 @@ Hi-Res Fixを接続しなければV3.4 Sampling経路は変わりません。強
|
||||
|
||||
詳しいsocket接続・OFF時のpassthrough・既知の制限は[英語版V3.5接続ガイド](docs/V35_HIRES_FIX.md)を参照してください。
|
||||
|
||||
## 推奨テンプレートワークフロー
|
||||
|
||||
- [V3.5推奨テンプレート](examples/workflows/MiniMax_H3_Continuum_V35.json)
|
||||
|
||||
V3.5テンプレートはHi-Res Fixが初期OFF、AssembleのBuffer BackendがAutoです。First Frame、Last Frame、Reference Image 1~3は1つのmegapixel設定を共有し、Video Referenceの解像度はVHS Load Video側で調整します。用途に応じて不要な補助経路を外したり、Advanced Second Pass経路へ組み替えたりできます。
|
||||
|
||||
## V3.4互換
|
||||
|
||||
V3.4ノードは保存済みワークフローを壊さないため残しています。Node ID、公開socket、Sampling、Conditioning、Terminal Merge、Assembly、Seam、Run StorageをV3.5へ置換していません。Second PassやDisk-backed Assembleを必要としない場合は、従来どおりV3.4を使用できます。
|
||||
|
||||
@@ -1,5 +1,13 @@
|
||||
# Work Log
|
||||
|
||||
## 2026-08-24 - V3.5 recommended template workflow
|
||||
|
||||
- Added `examples/workflows/MiniMax_H3_Continuum_V35.json` as the recommended V3.5 template while retaining every V3.4 template.
|
||||
- Verified the V3.5 Sampler -> optional Hi-Res Fix -> Core Video/Audio Decode -> Assemble + Seam V3.5 route, including `refine_context`, Video VAE, Reference Images 1-3, and Video Reference socket placement.
|
||||
- The template defaults Hi-Res Fix to OFF and Assemble to Auto. First/Last/Reference still images share one 0.3 MP control; VHS Video Reference sizing remains independent.
|
||||
- Updated English/Japanese README, Changelog, project state, and distribution Manifest. No Production Python or runtime file was changed.
|
||||
- Validation: JSON parse and semantic route audit PASS; all Manifest hashes PASS; full pytest `430 passed`; ComfyUI_W runtime registration, native PackedLayout, and Fixed 3x5 prompt-plan verification PASS; `git diff --check` PASS.
|
||||
|
||||
## 2026-08-22 - E2-8A / E2-8B1 implementation
|
||||
|
||||
- Added additive physical-group Second Pass metadata to V3.4 Assembly Plan output.
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user