我是个新手遇到一个无法解决的问题，模型都是在hugging face上下载的结果工作流无法运行

Open

#502 opened on May 26, 2026 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty: 4/5
Estimated time: 3-5 days
Newbie friendliness: 22/100

Issue type: Bug
Clarity: Needs clarification
Activity status: Quiet
Tech stack: python
Domain: machine-learning

Research direction

The payload gives no file paths, only the traceback: a state_dict size mismatch on LTXAVModel for scale_shift_table_a2v_ca_audio and scale_shift_table_a2v_ca_video across transformer_blocks. Start by finding where LTXAVModel is defined and where those tables are sized, then compare against the checkpoint the reporter downloaded from Hugging Face. Done means either the loader accepts the checkpoint or the user is pointed to the correct model version, confirmed by reproducing the workflow.

Written by the indexing model from the issue text.

Description

Error(s) in loading state_dict for LTXAVModel:
size mismatch for transformer_blocks.0.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.0.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.1.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.1.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.2.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.2.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.3.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.3.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.4.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.4.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.5.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.5.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.6.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.6.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.7.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.7.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.8.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.8.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.9.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.9.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.10.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.10.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.11.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.11.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.12.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.12.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.13.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.13.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.14.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.14.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.15.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.15.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.16.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.16.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.17.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.17.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.18.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.18.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.19.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.19.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.20.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.20.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.21.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.21.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.22.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.22.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.23.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.23.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.24.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.24.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.25.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.25.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.26.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.26.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.27.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.27.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.28.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.28.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.29.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.29.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.30.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.30.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.31.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.31.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.32.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.32.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.33.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.33.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.34.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.34.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.35.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.35.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.36.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.36.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.37.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.37.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.38.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.38.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.39.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.39.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.40.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.40.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.41.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.41.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.42.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.42.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.43.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.43.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.44.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.44.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.45.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.45.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.46.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.46.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).
size mismatch for transformer_blocks.47.scale_shift_table_a2v_ca_audio: copying a param with shape torch.Size([5, 2048]) from checkpoint, the shape in current model is torch.Size([8, 2048]).
size mismatch for transformer_blocks.47.scale_shift_table_a2v_ca_video: copying a param with shape torch.Size([5, 4096]) from checkpoint, the shape in current model is torch.Size([8, 4096]).

Dominant language: Python
Stars: 4.2k
Forks: 508
Avg merge: 7h 14m
Merged PRs (30d): 3

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

Read the whole issue, then the project's contributing guide.
Comment on the issue to say you are picking it up — it saves two people doing the same work.
Fork the repository and make your change on a branch.
Open a pull request that references the issue number.

Similar issues

[BUG] Elevated Group Discovery with PowerView: the test sample windows-xml.log is declared as Sysmon but holds Security events
Open
- Is something wrong?
bug

Difficulty 1/5 Under an hour Newbie friendliness 78/100

splunk/security_content#4418 · Opened Nov 14, 2026

Maintainers usually reply within 1 day
[Bug / Enhancement]: Improve teleop_robot CLI with configurable timeout, robust input validation, and clean ROS shutdown
Possibly taken @muskan-creates352 claimed this today. Open
- Is something wrong?
Difficulty 2/5 1-3 hours Newbie friendliness 72/100

open-rmf/rmf_demos#382 · Opened Nov 14, 2026

Maintainers usually reply within 3 days
bug(pipelines): shortcode stripping deletes text between a {{< >}} shortcode and the next {{% %}}
Possibly taken @Ashborne-lab claimed this today. Open
- Is something wrong?
Difficulty 2/5 1-3 hours Newbie friendliness 70/100

kubeflow/docs-agent#293 · Opened Nov 14, 2026

Maintainers usually reply within 3 days
citation_compliance_agent: some checks still assume APA 7th for every citation style
Open
- Is something wrong?
Difficulty 2/5 1-3 hours Newbie friendliness 62/100

Imbad0202/academic-research-skills#1093 · Opened Nov 14, 2026

Maintainers usually reply within 1 day
Input: placeholder drawn in the text colour when the empty Input is focused but not yet edited
Open
- Is something wrong?
Difficulty 2/5 1-3 hours Newbie friendliness 78/100

pierreaubert/gpui-toolkit#114 · Opened Nov 14, 2026

Maintainers usually reply within 1 day

More Python issues

我是个新手遇到一个无法解决的问题，模型都是在hugging face上下载的结果工作流无法运行

Is something wrong?

Assessment

Research direction

Description

Getting set up

First steps

More from Lightricks/ComfyUI-LTXVideo

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Similar issues

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Is something wrong?

Get new issues in your inbox