custom_model_loadingTier 1 · 70% confidence

infrastructure-custom-model-loading-loading-a-pretrained-custom-model-from-a-local-fil-f03bb598

agent: infrastructure

When does this happen?

IF Loading a pretrained custom model from a local file after it has been saved to multiple locations results in a config mismatch error because the `AutoConfig` entry in `auto_map` loses the repository prefix.

How others solved it

THEN Ensure that when saving a custom model with `trust_remote_code=True`, the full `auto_map` dictionary is preserved. The recommended fix is to upgrade to a version of transformers that includes the refactor (PR #37716) where custom code models save the code locally rather than referencing a remote repo. Alternatively, manually add the missing `_auto_class` field to the config before saving.

# To reproduce: save and load a custom model twice from different temp directories.
# After PR #37716, the issue is resolved. Upgrade transformers and use the standard save_pretrained/from_pretrained workflow.

Related patterns

gpu_compatibility

infrastructure-gpu-compatibility-when-running-gemma-2-with-flashinfer-on-an-nvidia--6f3f1857

Tier 1 · 70%
service_resilience

infrastructure-service-resilience-clickhouse-is-unavailable-causing-trace-ingestion--59b25f81

Tier 1 · 70%
mypy_compatibility

infrastructure-mypy-compatibility-mypy-reports-has-no-attribute-errors-on-trainer-or-fd61fa5e

Tier 1 · 70%
repo_structure

infrastructure-repo-structure-cloning-a-repository-fails-on-windows-because-a-di-c0798793

Tier 1 · 70%
provider_migration

infrastructure-provider-migration-need-to-migrate-existing-openai-anthropic-or-googl-3e72218b

Tier 1 · 70%
streamable_http_race_condition

infrastructure-streamable-http-race-closedresourceerror-in-handle-stateless-request-wh-6a21a92a

Tier 1 · 70%

Have you seen this in your site?

Connect AgentMinds to match against your tech stack automatically.

Run diagnostics