Motion Transfer v1.0, TARA v1.0 and Combined LoRAs + BFS Motion Transfer Community License 1.0
Browse files- LICENSES.md +17 -2
- LICENSE_MOTION_TRANSFER.md +50 -0
- README.md +8 -5
- loras/minimax_h3_motion_transfer_combined_v1.0_dyn95.safetensors +3 -0
- loras/minimax_h3_motion_transfer_combined_v1.0_dyn99.safetensors +3 -0
- loras/minimax_h3_motion_transfer_tara_v1.0_ref2va.safetensors +3 -0
- loras/minimax_h3_motion_transfer_v1.0_ref2va.safetensors +3 -0
LICENSES.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
# License information
|
| 2 |
|
| 3 |
Licenses in this repository can vary by asset. This file identifies the scope of the custom
|
| 4 |
-
Character Swap
|
| 5 |
|
| 6 |
## BFS Character Swap LoRAs
|
| 7 |
|
|
@@ -20,6 +20,21 @@ Organizations with consolidated annual gross revenue above US $1,000,000 must cr
|
|
| 20 |
and the BFS Character Swap LoRA in commercial use. The LoRA or its inference functionality may not
|
| 21 |
be offered as a hosted API, SaaS, or other remote model service. See the license for full terms.
|
| 22 |
|
| 23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
MiniMax H3 base-model files, or any other LoRA in this repository. Those assets retain their own
|
| 25 |
applicable licenses and upstream terms.
|
|
|
|
| 1 |
# License information
|
| 2 |
|
| 3 |
Licenses in this repository can vary by asset. This file identifies the scope of the custom
|
| 4 |
+
Character Swap and Motion Transfer licenses; it does not change the terms for other assets.
|
| 5 |
|
| 6 |
## BFS Character Swap LoRAs
|
| 7 |
|
|
|
|
| 20 |
and the BFS Character Swap LoRA in commercial use. The LoRA or its inference functionality may not
|
| 21 |
be offered as a hosted API, SaaS, or other remote model service. See the license for full terms.
|
| 22 |
|
| 23 |
+
## BFS Motion Transfer LoRAs
|
| 24 |
+
|
| 25 |
+
The [BFS Motion Transfer License](LICENSE_MOTION_TRANSFER.md) applies only to these weight files and derivative
|
| 26 |
+
weight files made from them:
|
| 27 |
+
|
| 28 |
+
- `loras/minimax_h3_motion_transfer_v1.0_ref2va.safetensors` — Motion Transfer v1.0 (guide latents).
|
| 29 |
+
- `loras/minimax_h3_motion_transfer_tara_v1.0_ref2va.safetensors` — Motion Transfer TARA v1.0 (native reference + TARA).
|
| 30 |
+
- `loras/minimax_h3_motion_transfer_combined_v1.0_dyn95.safetensors` and `..._dyn99.safetensors` — Motion Transfer
|
| 31 |
+
Combined (v1.0 × 1.0 + TARA v1.0 × 0.3, SVD-compressed; test).
|
| 32 |
+
|
| 33 |
+
They load on top of [ostris' training adapter V2](https://huggingface.co/ostris/minimax_h3_training_adapter), which keeps
|
| 34 |
+
its own terms. Same terms as Character Swap: free use including commercial videos and films; organizations above
|
| 35 |
+
US $1,000,000 annual gross revenue must credit Alissonerdx and the BFS Motion Transfer LoRA; no hosted API, SaaS or
|
| 36 |
+
remote model service.
|
| 37 |
+
|
| 38 |
+
These custom licenses do not cover the workflow JSON, documentation, example videos, ComfyUI or
|
| 39 |
MiniMax H3 base-model files, or any other LoRA in this repository. Those assets retain their own
|
| 40 |
applicable licenses and upstream terms.
|
LICENSE_MOTION_TRANSFER.md
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# BFS Motion Transfer Community License 1.0
|
| 2 |
+
|
| 3 |
+
Copyright (c) 2026 Alissonerdx
|
| 4 |
+
|
| 5 |
+
This license applies only to the Motion Transfer LoRA weight files listed in
|
| 6 |
+
[`LICENSES.md`](LICENSES.md) (Motion Transfer v1.0, Motion Transfer TARA v1.0 and the Motion Transfer Combined
|
| 7 |
+
merges) and derivative weight files made from any of them. It does not license MiniMax H3, ComfyUI, this repository's workflow,
|
| 8 |
+
documentation, example videos, or other models. Their respective terms continue to apply.
|
| 9 |
+
|
| 10 |
+
## 1. Permission
|
| 11 |
+
|
| 12 |
+
The copyright holder grants everyone a worldwide, non-exclusive, royalty-free permission, without
|
| 13 |
+
requiring prior approval, to download, use, reproduce, modify, and share the covered model weights,
|
| 14 |
+
subject to the conditions below. Personal, research, educational, and commercial use are allowed.
|
| 15 |
+
You may use the LoRA to create, publish, distribute, and sell videos, films, and other media.
|
| 16 |
+
|
| 17 |
+
If a company or other organization with consolidated annual gross revenue exceeding US $1,000,000
|
| 18 |
+
uses the covered weights or derivatives for a commercial purpose, it must give clear attribution to
|
| 19 |
+
the BFS Motion Transfer LoRA by Alissonerdx and link to the original model page:
|
| 20 |
+
https://huggingface.co/Alissonerdx/Minimax-H3-ComfyUI
|
| 21 |
+
Attribution may appear in the work's credits, description, or another reasonably visible place
|
| 22 |
+
associated with the work or service. Revenue is measured for the most recently completed fiscal year
|
| 23 |
+
and includes entities under common control. This revenue threshold does not prohibit commercial
|
| 24 |
+
use and does not require a separate permission or fee; it only triggers the attribution condition.
|
| 25 |
+
Independent creators and organizations are not treated as affiliates solely because they work with
|
| 26 |
+
or deliver work to a larger company.
|
| 27 |
+
|
| 28 |
+
## 2. Sharing
|
| 29 |
+
|
| 30 |
+
You may share the covered weights or derivative weights only if you include this license and retain
|
| 31 |
+
the copyright notice. Any recipient's use remains subject to this license. You may not sell or
|
| 32 |
+
repackage the covered weights, or charge for access to them, as a standalone product. You may not
|
| 33 |
+
offer the covered weights or their functionality as a hosted API, SaaS product, or other remote
|
| 34 |
+
inference service, whether paid or free. This restriction does not prevent using the LoRA internally
|
| 35 |
+
to create media, including commercial videos and films, or selling or distributing those resulting
|
| 36 |
+
media.
|
| 37 |
+
|
| 38 |
+
## 3. Other terms
|
| 39 |
+
|
| 40 |
+
This license does not grant rights to the MiniMax H3 base model, the training adapter V2 (ostris), the ComfyUI software or nodes, or
|
| 41 |
+
any third-party material. You are responsible for complying with those separate terms and for
|
| 42 |
+
having the rights and consent needed for your inputs and generated material.
|
| 43 |
+
|
| 44 |
+
The covered weights are provided "as is", without warranties or guarantees of any kind. To the
|
| 45 |
+
maximum extent permitted by applicable law, the copyright holder is not liable for claims or
|
| 46 |
+
damages arising from use of the covered weights.
|
| 47 |
+
|
| 48 |
+
This is a custom community license, not an OSI-approved Open Source license. The covered weights are
|
| 49 |
+
not offered for resale or as a hosted model service; the intent is to let people use the LoRA freely
|
| 50 |
+
while ensuring larger commercial uses credit its origin.
|
README.md
CHANGED
|
@@ -31,6 +31,9 @@ resolution, prompt and reference images.
|
|
| 31 |
| [`minimax_h3_head_swap_v1.0_r32`](loras/minimax_h3_head_swap_v1.0_r32.safetensors) | Experimental reference-image head replacement over an aligned source video | `head_swap:` | [Head Swap guide and prompt template](docs/head-swap.md) | [`minimax_h3_head_swap_workflow.json`](workflows/minimax_h3_head_swap_workflow.json) |
|
| 32 |
| [`minimax_h3_character_swap_v1.1_ref2va_fro`](loras/minimax_h3_character_swap_v1.1_ref2va_fro.safetensors) **(recommended)** | Replace the main guide subject with the character in a reference image while following the guide performance; 80 MB SVD extraction | `body_swap:` | [Character Swap guide, prompt template and examples](docs/character-swap.md) | [Shot Planner](workflows/minimax_h3_character_swap_workflow_v1.1.json) / [native nodes](workflows/minimax_h3_character_swap_workflow_native_nodes_v1.1.json) |
|
| 33 |
| [`minimax_h3_character_swap_v1.0_ref2va_r64_gl`](loras/minimax_h3_character_swap_v1.0_ref2va_r64_gl.safetensors) | Previous Character Swap release (rank 64) | `body_swap:` | [Character Swap guide](docs/character-swap.md) | [Shot Planner](workflows/minimax_h3_character_swap_workflow_v1.0.json) / [native nodes](workflows/minimax_h3_character_swap_workflow_native_nodes_v1.0.json) |
|
|
|
|
|
|
|
|
|
|
| 34 |
| [`minimax_h3_live_wallpaper_v1.0_ref2va_r32`](loras/minimax_h3_live_wallpaper_v1.0_ref2va_r32.safetensors) | Animate one to three reference images; often more faithful to facial features, but with less predictable camera behavior | `live_wallpaper:` | [Live Wallpaper guide](docs/live-wallpaper.md) | [`minimax_h3_live_wallpaper_workflow.json`](workflows/minimax_h3_live_wallpaper_workflow.json) |
|
| 35 |
| [`minimax_h3_live_wallpaper_v1.0_ref2va_r64`](loras/minimax_h3_live_wallpaper_v1.0_ref2va_r64.safetensors) | Camera-aware continuation of R32 for prompt-directed camera behavior and temporal reference frames | `live_wallpaper:` | [Live Wallpaper guide](docs/live-wallpaper.md) | [`minimax_h3_live_wallpaper_workflow.json`](workflows/minimax_h3_live_wallpaper_workflow.json) |
|
| 36 |
|
|
@@ -58,13 +61,13 @@ the guide shares the generated video's spatial and temporal grid.
|
|
| 58 |
|
| 59 |
## License and credits
|
| 60 |
|
| 61 |
-
Licenses are specified by asset. The separate [Character Swap license](LICENSE_CHARACTER_SWAP.md)
|
| 62 |
-
|
| 63 |
-
[LICENSES.md](LICENSES.md);
|
| 64 |
-
Use of
|
| 65 |
permission. Organizations with over US $1,000,000 in annual gross revenue must credit its origin
|
| 66 |
for commercial use. Offering the LoRA as a hosted API, SaaS, or other remote model service is not
|
| 67 |
-
allowed. See the
|
| 68 |
terms before use.
|
| 69 |
|
| 70 |
Base model credit to [MiniMaxAI](https://huggingface.co/MiniMaxAI/MiniMax-H3), ComfyUI-ready
|
|
|
|
| 31 |
| [`minimax_h3_head_swap_v1.0_r32`](loras/minimax_h3_head_swap_v1.0_r32.safetensors) | Experimental reference-image head replacement over an aligned source video | `head_swap:` | [Head Swap guide and prompt template](docs/head-swap.md) | [`minimax_h3_head_swap_workflow.json`](workflows/minimax_h3_head_swap_workflow.json) |
|
| 32 |
| [`minimax_h3_character_swap_v1.1_ref2va_fro`](loras/minimax_h3_character_swap_v1.1_ref2va_fro.safetensors) **(recommended)** | Replace the main guide subject with the character in a reference image while following the guide performance; 80 MB SVD extraction | `body_swap:` | [Character Swap guide, prompt template and examples](docs/character-swap.md) | [Shot Planner](workflows/minimax_h3_character_swap_workflow_v1.1.json) / [native nodes](workflows/minimax_h3_character_swap_workflow_native_nodes_v1.1.json) |
|
| 33 |
| [`minimax_h3_character_swap_v1.0_ref2va_r64_gl`](loras/minimax_h3_character_swap_v1.0_ref2va_r64_gl.safetensors) | Previous Character Swap release (rank 64) | `body_swap:` | [Character Swap guide](docs/character-swap.md) | [Shot Planner](workflows/minimax_h3_character_swap_workflow_v1.0.json) / [native nodes](workflows/minimax_h3_character_swap_workflow_native_nodes_v1.0.json) |
|
| 34 |
+
| [`minimax_h3_motion_transfer_v1.0_ref2va`](loras/minimax_h3_motion_transfer_v1.0_ref2va.safetensors) **(new)** | Motion Transfer: the character in the reference performs the guide video's motion; load after [ostris' training adapter V2](https://huggingface.co/ostris/minimax_h3_training_adapter) | `motion_transfer:` | [Henge](https://github.com/alisson-anjos/Henge) | Henge studio / [Motion Transfer license](LICENSE_MOTION_TRANSFER.md) |
|
| 35 |
+
| [`minimax_h3_motion_transfer_tara_v1.0_ref2va`](loras/minimax_h3_motion_transfer_tara_v1.0_ref2va.safetensors) **(new)** | Motion Transfer with the guide as a native video reference, aligned frame by frame by TARA (`BFSMiniMaxH3NativeTemporal`); with adapter V2 | `motion_transfer:` | [TARA examples](https://github.com/alisson-anjos/Henge/tree/main/workflows/examples) | [TARA graph](https://github.com/alisson-anjos/Henge/blob/main/workflows/examples/tara-native-10500-v2.comfy.json) |
|
| 36 |
+
| [`minimax_h3_motion_transfer_combined_v1.0_dyn95`](loras/minimax_h3_motion_transfer_combined_v1.0_dyn95.safetensors) / [`_dyn99`](loras/minimax_h3_motion_transfer_combined_v1.0_dyn99.safetensors) (test) | v1.0 × 1.0 + TARA v1.0 × 0.3 merged and SVD-compressed; with adapter V2 | `motion_transfer:` | [Henge](https://github.com/alisson-anjos/Henge) | Henge studio |
|
| 37 |
| [`minimax_h3_live_wallpaper_v1.0_ref2va_r32`](loras/minimax_h3_live_wallpaper_v1.0_ref2va_r32.safetensors) | Animate one to three reference images; often more faithful to facial features, but with less predictable camera behavior | `live_wallpaper:` | [Live Wallpaper guide](docs/live-wallpaper.md) | [`minimax_h3_live_wallpaper_workflow.json`](workflows/minimax_h3_live_wallpaper_workflow.json) |
|
| 38 |
| [`minimax_h3_live_wallpaper_v1.0_ref2va_r64`](loras/minimax_h3_live_wallpaper_v1.0_ref2va_r64.safetensors) | Camera-aware continuation of R32 for prompt-directed camera behavior and temporal reference frames | `live_wallpaper:` | [Live Wallpaper guide](docs/live-wallpaper.md) | [`minimax_h3_live_wallpaper_workflow.json`](workflows/minimax_h3_live_wallpaper_workflow.json) |
|
| 39 |
|
|
|
|
| 61 |
|
| 62 |
## License and credits
|
| 63 |
|
| 64 |
+
Licenses are specified by asset. The separate [Character Swap license](LICENSE_CHARACTER_SWAP.md) and
|
| 65 |
+
[Motion Transfer license](LICENSE_MOTION_TRANSFER.md) apply only to the Character Swap and Motion Transfer LoRA
|
| 66 |
+
weight files listed in [LICENSES.md](LICENSES.md); they do not apply to the other LoRAs, workflows, docs, or examples.
|
| 67 |
+
Use of these LoRAs, including commercial use in videos and films, requires no prior
|
| 68 |
permission. Organizations with over US $1,000,000 in annual gross revenue must credit its origin
|
| 69 |
for commercial use. Offering the LoRA as a hosted API, SaaS, or other remote model service is not
|
| 70 |
+
allowed. See the full [Character Swap](LICENSE_CHARACTER_SWAP.md) and [Motion Transfer](LICENSE_MOTION_TRANSFER.md) licenses and check each other asset's applicable
|
| 71 |
terms before use.
|
| 72 |
|
| 73 |
Base model credit to [MiniMaxAI](https://huggingface.co/MiniMaxAI/MiniMax-H3), ComfyUI-ready
|
loras/minimax_h3_motion_transfer_combined_v1.0_dyn95.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24c4c8180824cbe521d36447a79bcec75b0f7828ed973db9bd60b498d72d0b78
|
| 3 |
+
size 99138120
|
loras/minimax_h3_motion_transfer_combined_v1.0_dyn99.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ea6883fc11070a646f2f8e28261b6a59eb2caa5a9a75bf5a0b5f80a93b63335
|
| 3 |
+
size 199114376
|
loras/minimax_h3_motion_transfer_tara_v1.0_ref2va.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b00ca9e47f38f48638490c9a64059fffff839862f3d0c1f811a4e176cb05eaf
|
| 3 |
+
size 620286472
|
loras/minimax_h3_motion_transfer_v1.0_ref2va.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9146684b9c1412d235bd3943e7fddbf3e3aec8dc127cfea82c6aa07cc54f311d
|
| 3 |
+
size 620286016
|