diffusers
Add CogVideoX text-to-video generation model
#9082
Merged
Overview
Commits
107
Changes
View On GitHub

Add CogVideoX text-to-video generation model #9082

yiyixuxu merged 107 commits into huggingface:main from zRzRzRzRzRzRzR:cogvideox-2b
zRzRzRzRzRzRzR
zRzRzRzRzRzRzR Create autoencoder_kl3d.py
c8e5491b
zRzRzRzRzRzRzR vae draft
c341786f
a-r-r-o-w initial draft of cogvideo transformer
bd6efd5f
a-r-r-o-w add imports
bb917755
a-r-r-o-w fix attention mask
59e6669f
a-r-r-o-w fix layernorms
45cb1f92
zRzRzRzRzRzRzR fix with some review guide
84ff56eb
zRzRzRzRzRzRzR rename
a3d827fb
zRzRzRzRzRzRzR fix error
dc7e6e81
zRzRzRzRzRzRzR Update autoencoder_kl3d.py
aff72ec5
a-r-r-o-w fix nasty bug in 3d sincos pos embeds
cb5348a0
a-r-r-o-w refactor
e9828817
a-r-r-o-w update conversion script for latest modeling changes
d963b1aa
a-r-r-o-w remove debug prints
16967589
a-r-r-o-w make style
21a0fc1b
sayakpaul add workflow to rebase with upstream main nightly.
d83c1f84
sayakpaul add upstream
dfeb3297
sayakpaul Revert "add workflow to rebase with upstream main nightly."
71bcb1e1
sayakpaul add workflow for rebasing with upstream automatically.
0980f4dc
zRzRzRzRzRzRzR follow review guide
ee40f0e1
zRzRzRzRzRzRzR add
8fe54bcd
zRzRzRzRzRzRzR remove deriving and using nn.module
1c661ce3
a-r-r-o-w Merge branch 'cogvideox' into cogvideox-common-draft-1
73b041e7
a-r-r-o-w add skeleton for pipeline
b3052807
a-r-r-o-w make fix-copies
6bcafcba
a-r-r-o-w Merge branch 'main' into cogvideox-common-draft-2
ec9508c8
a-r-r-o-w undo unnecessary changes added on cogvideo-vae by mistake
3ae94139
a-r-r-o-w groups->norm_num_groups
2be74698
a-r-r-o-w verify CogVideoXSpatialNorm3D implementation
9f9d0cbb
a-r-r-o-w minor factor and repositioning of code in order of invocation
c43a8f5b
a-r-r-o-w reorder upsampling/downsampling blocks in order of invocation
5f183bfe
a-r-r-o-w minor refactor
470815ce
a-r-r-o-w implement encode prompt
e67cc5ae
a-r-r-o-w make style
d45d199b
a-r-r-o-w make fix-copies
73469f95
a-r-r-o-w fix bug in handling long prompts
45f7127a
a-r-r-o-w update conversion script
a449ceb3
zRzRzRzRzRzRzR add doc draft
4498cfc9
zRzRzRzRzRzRzR Merge branch 'cogvideox-common-draft-2' of https://github.com/hugging…
2956866e
zRzRzRzRzRzRzR add clear_fake_cp_cache
bb4740ce
a-r-r-o-w refactor vae
e05f8347
a-r-r-o-w modeling fixes
03c28eef
a-r-r-o-w make style
712ddbea
a-r-r-o-w add pipeline implementation
03ee7cd1
zRzRzRzRzRzRzR using with 226 instead of 225 of final weight
a31db5f9
zRzRzRzRzRzRzR remove 0.transformer_blocks.encoder.embed_tokens.weight
351d1f00
a-r-r-o-w update
d0b8db2b
a-r-r-o-w ensure tokenizer config correctly uses 226 as text length
fe6f5d64
a-r-r-o-w add cogvideo specific attn processor
4c2e8870
a-r-r-o-w remove debug prints
41da084f
a-r-r-o-w add pipeline docs
77558f31
a-r-r-o-w make style
e12458e1
a-r-r-o-w remove incorrect copied from
c33dd021
zRzRzRzRzRzRzR vae problem fix
71e7c82a
zRzRzRzRzRzRzR schedule
ec53a30a
a-r-r-o-w remove debug prints
551c884a
a-r-r-o-w update
3def9052
a-r-r-o-w Merge pull request #4 from huggingface/cogvideox-refactor-to-diffusers
65f6211f
zRzRzRzRzRzRzR fp16 problem
21509aa7
zRzRzRzRzRzRzR fix some comment
b42b0792
zRzRzRzRzRzRzR fix
477e12b2
zRzRzRzRzRzRzR timestep fix
fd0831c5
zRzRzRzRzRzRzR Restore the timesteps parameter
d99528be
zRzRzRzRzRzRzR Update downsampling.py
c7ee165c
a-r-r-o-w remove chunked ff code; reuse and refactor to support temb directly i…
61c6da07
a-r-r-o-w make inference 2-3x faster (by fixing the bug i introduced) 🚀😎
fa7fa9cc
zRzRzRzRzRzRzR new schedule with dpm
6988cc3a
zRzRzRzRzRzRzR remove attenstion mask
ba4223ac
a-r-r-o-w apply suggestions from review
312f7dc4
a-r-r-o-w make style
1b1b26b6
sayakpaul add workflow to rebase with upstream main nightly.
ba1855c0
sayakpaul add upstream
7360ea1d
sayakpaul Revert "add workflow to rebase with upstream main nightly."
2f1b7870
sayakpaul add workflow for rebasing with upstream automatically.
90aa8be5
a-r-r-o-w Merge branch 'huggingface:main' into main
5781e017
a-r-r-o-w make fix-copies
92c8c007
a-r-r-o-w Merge branch 'main' into cogvideox-common-draft-2
fd11c0fb
a-r-r-o-w remove cogvideox-specific attention processor
03580c07
a-r-r-o-w update docs
01c2dff3
a-r-r-o-w update docs
311845fc
zRzRzRzRzRzRzR cogvideox branch
1b1b737a
zRzRzRzRzRzRzR add CogVideoX team, Tsinghua University & ZhipuAI
2d9602cc
zRzRzRzRzRzRzR Merge branch 'cogvideox-common-draft-2' of github.com:huggingface/dif…
fb6130fe
zRzRzRzRzRzRzR merge remote branch
511c9ef5
zRzRzRzRzRzRzR zRzRzRzRzRzRzR changed the title Cogvideox 2b Add CogVideoX text-to-video generation model 1 year ago
a-r-r-o-w
a-r-r-o-w commented on 2024-08-05
a-r-r-o-w
a-r-r-o-w commented on 2024-08-05
a-r-r-o-w Merge branch 'main' into cogvideox-2b
123ecef2
DN6
DN6 commented on 2024-08-05
DN6
DN6 commented on 2024-08-05
DN6
DN6 commented on 2024-08-05
DN6
DN6 commented on 2024-08-05
DN6
DN6 commented on 2024-08-05
zRzRzRzRzRzRzR fix some error
cf7369d4
DN6
DN6 commented on 2024-08-05
zRzRzRzRzRzRzR rename unsample and add some docs
9c6b8894
zRzRzRzRzRzRzR messages
22dcceb8
yiyixuxu update
e4d65ccd
yiyixuxu Merge branch 'cogvideox-2b' of github.com:zRzRzRzRzRzRzR/diffusers in…
6f4e60b5
a-r-r-o-w use num_frames instead of num_seconds
70a54a82
a-r-r-o-w Merge branch 'main' into cogvideox-2b
b3428ad5
a-r-r-o-w
a-r-r-o-w commented on 2024-08-05
zRzRzRzRzRzRzR restore
9a0b9065
zRzRzRzRzRzRzR Update lora_conversion_utils.py
32da2e76
a-r-r-o-w remove dynamic guidance scale
878f609a
yiyixuxu
yiyixuxu commented on 2024-08-05
yiyixuxu
yiyixuxu
yiyixuxu
yiyixuxu
yiyixuxu
yiyixuxu commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
sayakpaul
sayakpaul commented on 2024-08-06
a-r-r-o-w address review comments
de9e0b2f
a-r-r-o-w dynamic cfg; fix cfg support
9c086f5a
a-r-r-o-w address review comments
62d94aaa
a-r-r-o-w update tests
5e4dd151
a-r-r-o-w Merge branch 'main' into cogvideox-2b
884ddd09
a-r-r-o-w fix docs error
d1c575ad
HuggingFaceDocBuilderDev
sayakpaul
sayakpaul commented on 2024-08-06
a-r-r-o-w alternative implementation to context parallel cache
11224d95
a-r-r-o-w
a-r-r-o-w
a-r-r-o-w commented on 2024-08-06
yiyixuxu
yiyixuxu approved these changes on 2024-08-06
yiyixuxu Update docs/source/en/api/pipelines/cogvideox.md
70cea915
stevhliu
stevhliu approved these changes on 2024-08-06
a-r-r-o-w remove tiling and slicing until their implementations are complete
cbc4d32d
yiyixuxu
yiyixuxu
yiyixuxu commented on 2024-08-06
sayakpaul Merge branch 'main' into cogvideox-2b
14698d04
sayakpaul Merge branch 'main' into cogvideox-2b
8be845d3
sayakpaul Apply suggestions from code review
827a70ae
sayakpaul
sayakpaul commented on 2024-08-07
sayakpaul
sayakpaul commented on 2024-08-07
sayakpaul
sayakpaul commented on 2024-08-07
yiyixuxu yiyixuxu merged 2dad462d into main 1 year ago
hkunzhe
hkunzhe commented on 2024-10-18
zRzRzRzRzRzRzR zRzRzRzRzRzRzR deleted the cogvideox-2b branch 259 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum