Skip to content

Add safetensors checkpoint loading support - #599

Open
johnwbrisbin wants to merge 1 commit into
facebookresearch:sam3.1from
johnwbrisbin:fix/safetensors-checkpoint-loading
Open

johnwbrisbin wants to merge 1 commit into
facebookresearch:sam3.1from
johnwbrisbin:fix/safetensors-checkpoint-loading

Conversation

@johnwbrisbin

Copy link
Copy Markdown

Summary

All four checkpoint-loading paths now detect a .safetensors extension and use safetensors.torch.load_file instead of torch.load:

  • _load_checkpoint helper
  • build_sam3_video_model
  • build_sam3_multiplex_video_model
  • build_sam3_multiplex_video_predictor

Motivation

The official Comfy-Org/sam3.1 HuggingFace distribution ships sam3.1_multiplex_fp16.safetensors. Passing this file to any of the above builders fails with:

ValueError: unsupported pickle protocol: 104

The safetensors magic bytes are misread as a legacy pickle header by torch.load.

.pt / .pth checkpoints continue to use torch.load, now with weights_only=False so that legacy checkpoints using older pickle protocols still load without raising UnpicklingError.

Test plan

  • Load sam3.1_multiplex_fp16.safetensors via build_sam3_multiplex_video_model — previously crashed, now loads correctly
  • Load a .pt checkpoint via build_sam3_video_model — still loads correctly with weights_only=False

All four checkpoint-loading paths (build_sam3_video_model,
build_sam3_multiplex_video_model, build_sam3_multiplex_video_predictor,
and the internal _load_checkpoint helper) now detect a .safetensors
extension and use safetensors.torch.load_file instead of torch.load.

The official Comfy-Org/sam3.1 HuggingFace distribution ships
sam3.1_multiplex_fp16.safetensors; torch.load fails on this format with
"unsupported pickle protocol: 104" (the safetensors magic bytes are
misread as a pickle header).

.pt / .pth checkpoints continue to use torch.load, now with
weights_only=False so that legacy checkpoints using older pickle
protocols still load without raising UnpicklingError.
@meta-cla

meta-cla Bot commented Jul 1, 2026

Copy link
Copy Markdown

Hi @johnwbrisbin!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!

@meta-cla meta-cla Bot added the CLA Signed This label is managed by the Meta Open Source bot. label Jul 1, 2026
@meta-cla

meta-cla Bot commented Jul 1, 2026

Copy link
Copy Markdown

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Meta Open Source bot.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant