Skip to content

[LTX2] Fix from_single_file: missing 2.3 key mappings and shared checkpoint dict draining #14151

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

Open
Functionhx wants to merge 1 commit into
base: main
Choose a base branch
from
+17 −3
Open

[LTX2] Fix from_single_file: missing 2.3 key mappings and shared checkpoint dict draining #14151

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
20 changes: 17 additions & 3 deletions src/diffusers/loaders/single_file_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4048,6 +4048,18 @@ def convert_ltx2_transformer_adaln_single(key: str, state_dict) -> None:
if ".weight" not in key and ".bias" not in key:
return

if key.startswith("prompt_adaln_single."):
new_key = key.replace("prompt_adaln_single.", "prompt_adaln.")
param = state_dict.pop(key)
state_dict[new_key] = param
return

if key.startswith("audio_prompt_adaln_single."):
new_key = key.replace("audio_prompt_adaln_single.", "audio_prompt_adaln.")
param = state_dict.pop(key)
state_dict[new_key] = param
return

if key.startswith("adaln_single."):
new_key = key.replace("adaln_single.", "time_embed.")
param = state_dict.pop(key)
Expand All @@ -4066,7 +4078,7 @@ def convert_ltx2_transformer_adaln_single(key: str, state_dict) -> None:
"adaln_single": convert_ltx2_transformer_adaln_single,
}

converted_state_dict = {key: checkpoint.pop(key) for key in list(checkpoint.keys())}
converted_state_dict = {key: checkpoint.pop(key) for key in list(checkpoint.keys()) if key.startswith("model.diffusion_model.")}

# Handle official code --> diffusers key remapping via the remap dict
for key in list(converted_state_dict.keys()):
Expand Down Expand Up @@ -4109,6 +4121,8 @@ def convert_ltx2_vae_to_diffusers(checkpoint, **kwargs):
"up_blocks.4": "up_blocks.1",
"up_blocks.5": "up_blocks.2.upsamplers.0",
"up_blocks.6": "up_blocks.2",
"up_blocks.7": "up_blocks.3.upsamplers.0",
"up_blocks.8": "up_blocks.3",
# Common
# For all 3D ResNets
"res_blocks": "resnets",
Expand All @@ -4127,7 +4141,7 @@ def remove_keys_inplace(key: str, state_dict) -> None:
"per_channel_statistics.mean-of-stds": remove_keys_inplace,
}

converted_state_dict = {key: checkpoint.pop(key) for key in list(checkpoint.keys())}
converted_state_dict = {key: checkpoint.pop(key) for key in list(checkpoint.keys()) if key.startswith("vae.")}

# Handle official code --> diffusers key remapping via the remap dict
for key in list(converted_state_dict.keys()):
Expand Down Expand Up @@ -4159,7 +4173,7 @@ def convert_ltx2_audio_vae_to_diffusers(checkpoint, **kwargs):
def update_state_dict_inplace(state_dict, old_key: str, new_key: str) -> None:
state_dict[new_key] = state_dict.pop(old_key)

converted_state_dict = {key: checkpoint.pop(key) for key in list(checkpoint.keys())}
converted_state_dict = {key: checkpoint.pop(key) for key in list(checkpoint.keys()) if key.startswith("audio_vae.")}

# Handle official code --> diffusers key remapping via the remap dict
for key in list(converted_state_dict.keys()):
Expand Down
Loading