Skip to content

Hunyuanvideo15 #12696

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yiyixuxu merged 35 commits into from
Dec 1, 2025
Merged

Hunyuanvideo15 #12696

Show file tree
Hide file tree
Changes from 15 commits
Commits
Show all changes
35 commits
Select commit Hold shift + click to select a range
8b7ea81
add
yiyixuxu Nov 20, 2025
56d57c3
add first pipeline draft
yiyixuxu Nov 22, 2025
b282ac1
add text encoders to conversion script
yiyixuxu Nov 22, 2025
5732d60
fix a bit more, remove print lines
yiyixuxu Nov 22, 2025
76bb607
fix more, system prompt etc
yiyixuxu Nov 23, 2025
c739ee9
update conversion script
Nov 26, 2025
2f6914d
up up
Nov 26, 2025
a0b2fe0
update conversion script: remove dtype, always keep same precision as…
Nov 27, 2025
db0127c
fix
Nov 27, 2025
38c42b4
conversion scripts
Nov 27, 2025
090ceb5
remove dtype from the _get_ encodeing methods
Nov 29, 2025
e3301cb
add i2v pipeline
Nov 29, 2025
753d407
add image to video pipeline
Nov 29, 2025
0687a40
remove use_meanflow
Nov 29, 2025
c22915d
up up
Nov 29, 2025
f9cb82b
a few small fix: proprocess, cpu_offloading, attention backend
Nov 30, 2025
e319d72
simplify transformer
Nov 30, 2025
e194034
clean up a bit more pipelines
Nov 30, 2025
3980f97
Merge branch 'main' into hunyuanvideo15
yiyixuxu Nov 30, 2025
5029dbf
style
Nov 30, 2025
8aa458e
copies
Nov 30, 2025
50abf50
add docs
Nov 30, 2025
7aeab3f
add to toctree
Nov 30, 2025
c3f4598
up
Nov 30, 2025
54f008e
Update docs/source/en/api/pipelines/hunyuan_video15.md
yiyixuxu Nov 30, 2025
237d318
Apply suggestions from code review
yiyixuxu Nov 30, 2025
d7f399d
add a notes on the doc about attention backend
Dec 1, 2025
bdfab30
up
Dec 1, 2025
2c018f8
Update docs/source/en/api/pipelines/hunyuan_video15.md
yiyixuxu Dec 1, 2025
c715470
add a note on changing guidance_scale on doc
yiyixuxu Dec 1, 2025
0dae8f9
Apply suggestions from code review
yiyixuxu Dec 1, 2025
5989014
Update docs/source/en/api/pipelines/hunyuan_video15.md
yiyixuxu Dec 1, 2025
404d3fa
Update docs/source/en/api/pipelines/hunyuan_video15.md
yiyixuxu Dec 1, 2025
0869b22
tests for Hunyuan 1.5 (#12759)
sayakpaul Dec 1, 2025
6bfb75a
Apply style fixes
github-actions[bot] Dec 1, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
845 changes: 845 additions & 0 deletions scripts/convert_hunyuan_video1_5_to_diffusers.py
View file
Open in desktop

Large diffs are not rendered by default.

8 changes: 8 additions & 0 deletions src/diffusers/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -189,6 +189,7 @@
"AutoencoderKLHunyuanImage",
"AutoencoderKLHunyuanImageRefiner",
"AutoencoderKLHunyuanVideo",
"AutoencoderKLHunyuanVideo15",
"AutoencoderKLLTXVideo",
"AutoencoderKLMagvit",
"AutoencoderKLMochi",
Expand Down Expand Up @@ -223,6 +224,7 @@
"HunyuanDiT2DModel",
"HunyuanDiT2DMultiControlNetModel",
"HunyuanImageTransformer2DModel",
"HunyuanVideo15Transformer3DModel",
"HunyuanVideoFramepackTransformer3DModel",
"HunyuanVideoTransformer3DModel",
"I2VGenXLUNet",
Expand Down Expand Up @@ -480,6 +482,8 @@
"HunyuanVideoFramepackPipeline",
"HunyuanVideoImageToVideoPipeline",
"HunyuanVideoPipeline",
"HunyuanVideo15Pipeline",
"HunyuanVideo15ImageToVideoPipeline",
"I2VGenXLPipeline",
"IFImg2ImgPipeline",
"IFImg2ImgSuperResolutionPipeline",
Expand Down Expand Up @@ -903,6 +907,7 @@
AutoencoderKLHunyuanImage,
AutoencoderKLHunyuanImageRefiner,
AutoencoderKLHunyuanVideo,
AutoencoderKLHunyuanVideo15,
AutoencoderKLLTXVideo,
AutoencoderKLMagvit,
AutoencoderKLMochi,
Expand Down Expand Up @@ -939,6 +944,7 @@
HunyuanImageTransformer2DModel,
HunyuanVideoFramepackTransformer3DModel,
HunyuanVideoTransformer3DModel,
HunyuanVideo15Transformer3DModel,
I2VGenXLUNet,
Kandinsky3UNet,
Kandinsky5Transformer3DModel,
Expand Down Expand Up @@ -1164,6 +1170,8 @@
HunyuanVideoFramepackPipeline,
HunyuanVideoImageToVideoPipeline,
HunyuanVideoPipeline,
HunyuanVideo15Pipeline,
HunyuanVideo15ImageToVideoPipeline,
I2VGenXLPipeline,
IFImg2ImgPipeline,
IFImg2ImgSuperResolutionPipeline,
Expand Down
4 changes: 4 additions & 0 deletions src/diffusers/models/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,7 @@
_import_structure["autoencoders.autoencoder_kl_hunyuan_video"] = ["AutoencoderKLHunyuanVideo"]
_import_structure["autoencoders.autoencoder_kl_hunyuanimage"] = ["AutoencoderKLHunyuanImage"]
_import_structure["autoencoders.autoencoder_kl_hunyuanimage_refiner"] = ["AutoencoderKLHunyuanImageRefiner"]
_import_structure["autoencoders.autoencoder_kl_hunyuanvideo15"] = ["AutoencoderKLHunyuanVideo15"]
_import_structure["autoencoders.autoencoder_kl_ltx"] = ["AutoencoderKLLTXVideo"]
_import_structure["autoencoders.autoencoder_kl_magvit"] = ["AutoencoderKLMagvit"]
_import_structure["autoencoders.autoencoder_kl_mochi"] = ["AutoencoderKLMochi"]
Expand Down Expand Up @@ -83,6 +84,7 @@
_import_structure["transformers.t5_film_transformer"] = ["T5FilmDecoder"]
_import_structure["transformers.transformer_2d"] = ["Transformer2DModel"]
_import_structure["transformers.transformer_allegro"] = ["AllegroTransformer3DModel"]
_import_structure["transformers.transformer_hunyuan_video15"] = ["HunyuanVideo15Transformer3DModel"]
_import_structure["transformers.transformer_bria"] = ["BriaTransformer2DModel"]
_import_structure["transformers.transformer_bria_fibo"] = ["BriaFiboTransformer2DModel"]
_import_structure["transformers.transformer_chroma"] = ["ChromaTransformer2DModel"]
Expand Down Expand Up @@ -143,6 +145,7 @@
AutoencoderKLHunyuanImage,
AutoencoderKLHunyuanImageRefiner,
AutoencoderKLHunyuanVideo,
AutoencoderKLHunyuanVideo15,
AutoencoderKLLTXVideo,
AutoencoderKLMagvit,
AutoencoderKLMochi,
Expand Down Expand Up @@ -191,6 +194,7 @@
DualTransformer2DModel,
EasyAnimateTransformer3DModel,
FluxTransformer2DModel,
HunyuanVideo15Transformer3DModel,
HiDreamImageTransformer2DModel,
HunyuanDiT2DModel,
HunyuanImageTransformer2DModel,
Expand Down
1 change: 1 addition & 0 deletions src/diffusers/models/autoencoders/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@
from .autoencoder_kl_hunyuan_video import AutoencoderKLHunyuanVideo
from .autoencoder_kl_hunyuanimage import AutoencoderKLHunyuanImage
from .autoencoder_kl_hunyuanimage_refiner import AutoencoderKLHunyuanImageRefiner
from .autoencoder_kl_hunyuanvideo15 import AutoencoderKLHunyuanVideo15
from .autoencoder_kl_ltx import AutoencoderKLLTXVideo
from .autoencoder_kl_magvit import AutoencoderKLMagvit
from .autoencoder_kl_mochi import AutoencoderKLMochi
Expand Down
Loading