Impact
A trust_remote_code bypass in DiffusionPipeline.from_pretrained allows arbitrary remote code execution despite the user passing trust_remote_code=False (or omitting it, which is the default). The vulnerability has three variants, all sharing the same root cause — the trust_remote_code gate was implemented inside DiffusionPipeline.download() rather than at the actual dynamic-module load site, so any code path that bypassed or short-circuited download() also bypassed the security check:
- Cross-repo
custom_pipeline. DiffusionPipeline.from_pretrained('repoA', custom_pipeline='attacker/repoB', trust_remote_code=False) — the gate evaluated against repoA's file list rather than repoB's, so repoB's pipeline.py was loaded and executed.
- Local snapshot + Hub
custom_pipeline. DiffusionPipeline.from_pretrained('/local/snapshot', custom_pipeline='attacker/repoB', trust_remote_code=False) — the local-path branch never invoked download(), so the gate was never reached and remote code from repoB executed.
- Local snapshot with custom components.
DiffusionPipeline.from_pretrained('/local/snapshot', trust_remote_code=False) where the snapshot contains custom component files (e.g. unet/my_unet_model.py) referenced from model_index.json — same root cause; the local path skipped download() and custom component code executed.
Silent remote code execution on the victim's machine. Anyone calling DiffusionPipeline.from_pretrained with custom pipelines is impacted.
Patches
Yes. Fixed in diffusers 0.38.0 via PR #13448. All users on versions < 0.38.0 should upgrade:
pip install --upgrade "diffusers>=0.38.0"
The fix moves the trust_remote_code gate out of DiffusionPipeline.download() and into get_cached_module_file in src/diffusers/utils/dynamic_modules_utils.py, which is the actual chokepoint for every dynamic module load (local, Hub, or community mirror). All three variants now raise ValueError instead of executing untrusted code.
Workarounds
If upgrading immediately is not possible:
- Only call
from_pretrained with pretrained_model_name_or_path, custom_pipeline, and local snapshot directories from fully trusted sources that have been audited.
- Do not pass
custom_pipeline= pointing at a Hub repository different from the primary pretrained_model_name_or_path before reading its pipeline.py.
- Before calling
from_pretrained on a local snapshot, inspect the snapshot for unexpected *.py files, especially under component subdirectories (unet/, scheduler/, etc.) and at the snapshot root.
These are mitigations, not fixes — the only complete remediation is upgrading to 0.38.0.
Resources
References
Impact
A
trust_remote_codebypass inDiffusionPipeline.from_pretrainedallows arbitrary remote code execution despite the user passingtrust_remote_code=False(or omitting it, which is the default). The vulnerability has three variants, all sharing the same root cause — thetrust_remote_codegate was implemented insideDiffusionPipeline.download()rather than at the actual dynamic-module load site, so any code path that bypassed or short-circuiteddownload()also bypassed the security check:custom_pipeline.DiffusionPipeline.from_pretrained('repoA', custom_pipeline='attacker/repoB', trust_remote_code=False)— the gate evaluated againstrepoA's file list rather thanrepoB's, sorepoB'spipeline.pywas loaded and executed.custom_pipeline.DiffusionPipeline.from_pretrained('/local/snapshot', custom_pipeline='attacker/repoB', trust_remote_code=False)— the local-path branch never invokeddownload(), so the gate was never reached and remote code fromrepoBexecuted.DiffusionPipeline.from_pretrained('/local/snapshot', trust_remote_code=False)where the snapshot contains custom component files (e.g.unet/my_unet_model.py) referenced frommodel_index.json— same root cause; the local path skippeddownload()and custom component code executed.Silent remote code execution on the victim's machine. Anyone calling
DiffusionPipeline.from_pretrainedwith custom pipelines is impacted.Patches
Yes. Fixed in diffusers 0.38.0 via PR #13448. All users on versions
< 0.38.0should upgrade:pip install --upgrade "diffusers>=0.38.0"The fix moves the
trust_remote_codegate out ofDiffusionPipeline.download()and intoget_cached_module_fileinsrc/diffusers/utils/dynamic_modules_utils.py, which is the actual chokepoint for every dynamic module load (local, Hub, or community mirror). All three variants now raiseValueErrorinstead of executing untrusted code.Workarounds
If upgrading immediately is not possible:
from_pretrainedwithpretrained_model_name_or_path,custom_pipeline, and local snapshot directories from fully trusted sources that have been audited.custom_pipeline=pointing at a Hub repository different from the primarypretrained_model_name_or_pathbefore reading itspipeline.py.from_pretrainedon a local snapshot, inspect the snapshot for unexpected*.pyfiles, especially under component subdirectories (unet/,scheduler/, etc.) and at the snapshot root.These are mitigations, not fixes — the only complete remediation is upgrading to 0.38.0.
Resources
References