Skip to content

fix: bug: ModelBuilder overwrites user-provided HF_MODEL_ID for DJL Serving, preventi (5529) #5734

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

Merged
aviruthen merged 7 commits into from
Apr 17, 2026
+390 −8
Merged

fix: bug: ModelBuilder overwrites user-provided HF_MODEL_ID for DJL Serving, preventi (5529) #5734

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
228240a
fix: bug: ModelBuilder overwrites user-provided HF_MODEL_ID for DJL S…
aviruthen Apr 7, 2026
4badbdd
fix: address review comments (iteration #1)
aviruthen Apr 7, 2026
22e4363
fix: address review comments (iteration #1)
aviruthen Apr 14, 2026
ab7b3c8
fix: address review comments (iteration #1)
aviruthen Apr 16, 2026
3050387
fix: address review comments (iteration #2)
aviruthen Apr 16, 2026
603d9f3
Merge branch 'master' into fix/bug-modelbuilder-overwrites-user-provi…
aviruthen Apr 16, 2026
ad0776f
Merge branch 'master' into fix/bug-modelbuilder-overwrites-user-provi…
aviruthen Apr 17, 2026
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
16 changes: 8 additions & 8 deletions sagemaker-serve/src/sagemaker/serve/model_builder_servers.py
View file
Open in desktop
Comment thread
aviruthen marked this conversation as resolved.
Comment thread
aviruthen marked this conversation as resolved.
Original file line number Diff line number Diff line change
Expand Up @@ -136,7 +136,7 @@ def _build_for_torchserve(self) -> Model:
if isinstance(self.model, str):
# Configure HuggingFace model support
if not self._is_jumpstart_model_id():
self.env_vars.update({"HF_MODEL_ID": self.model})
self.env_vars.setdefault("HF_MODEL_ID", self.model)
Comment thread
aviruthen marked this conversation as resolved.

# Add HuggingFace token if available
if self.env_vars.get("HUGGING_FACE_HUB_TOKEN"):
Expand Down Expand Up @@ -212,7 +212,7 @@ def _build_for_tgi(self) -> Model:

if isinstance(self.model, str) and not self._is_jumpstart_model_id():
# Configure HuggingFace model for TGI
self.env_vars.update({"HF_MODEL_ID": self.model})
self.env_vars.setdefault("HF_MODEL_ID", self.model)

self.hf_model_config = _get_model_config_properties_from_hf(
self.model, self.env_vars.get("HUGGING_FACE_HUB_TOKEN")
Expand Down Expand Up @@ -319,9 +319,9 @@ def _build_for_djl(self) -> Model:
logger.debug(f"Using detected notebook instance type: {nb_instance}")

if isinstance(self.model, str) and not self._is_jumpstart_model_id():
# Configure HuggingFace model for DJL (preserve user-provided HF_MODEL_ID)
# Configure HuggingFace model for DJL
self.env_vars.setdefault("HF_MODEL_ID", self.model)

# Get model configuration for DJL optimization
self.hf_model_config = _get_model_config_properties_from_hf(
self.model, self.env_vars.get("HUGGING_FACE_HUB_TOKEN")
Expand Down Expand Up @@ -432,7 +432,7 @@ def _build_for_triton(self) -> Model:
self.env_vars.update({"HF_TASK": model_task})

# Configure HuggingFace authentication
self.env_vars.update({"HF_MODEL_ID": self.model})
self.env_vars.setdefault("HF_MODEL_ID", self.model)
if self.env_vars.get("HUGGING_FACE_HUB_TOKEN"):
self.env_vars["HF_TOKEN"] = self.env_vars.get("HUGGING_FACE_HUB_TOKEN")

Expand Down Expand Up @@ -538,7 +538,7 @@ def _build_for_tei(self) -> Model:

if isinstance(self.model, str) and not self._is_jumpstart_model_id():
# Configure HuggingFace model for TEI
self.env_vars.update({"HF_MODEL_ID": self.model})
self.env_vars.setdefault("HF_MODEL_ID", self.model)

self.hf_model_config = _get_model_config_properties_from_hf(
self.model, self.env_vars.get("HUGGING_FACE_HUB_TOKEN")
Expand Down Expand Up @@ -682,7 +682,7 @@ def _build_for_transformers(self) -> Model:
if self.inference_spec is not None:
hf_model_id = self.inference_spec.get_model()
if isinstance(hf_model_id, str): # Only if it's a valid HF model ID
self.env_vars.update({"HF_MODEL_ID": hf_model_id})
self.env_vars.setdefault("HF_MODEL_ID", hf_model_id)
# Get HF config only for string model IDs
if hasattr(self.env_vars, "HF_API_TOKEN"):
self.hf_model_config = _get_model_config_properties_from_hf(
Expand All @@ -701,7 +701,7 @@ def _build_for_transformers(self) -> Model:
if model_task:
self.env_vars.update({"HF_TASK": model_task})

self.env_vars.update({"HF_MODEL_ID": self.model})
self.env_vars.setdefault("HF_MODEL_ID", self.model)

# Add HuggingFace token if available
if self.env_vars.get("HUGGING_FACE_HUB_TOKEN"):
Expand Down
Loading
Loading