From 5d52e0d2c5c4550eceeabed5115cf552295819db Mon Sep 17 00:00:00 2001
From: ehhuang <ehhuang@users.noreply.github.com>
Date: Thu, 4 Sep 2025 04:23:18 -0700
Subject: [PATCH 1/3] chore: handle missing finish_reason (#3328)

# What does this PR do?
Sometimes the stream don't have chunks with finish_reason, e.g. canceled
stream, which throws a pydantic error as OpenAIChoice.finish_reason: str

## Test Plan
observe no more such error when benchmarking
---
 llama_stack/core/routers/inference.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/llama_stack/core/routers/inference.py b/llama_stack/core/routers/inference.py
index 4b66601bb..8dcad85e3 100644
--- a/llama_stack/core/routers/inference.py
+++ b/llama_stack/core/routers/inference.py
@@ -755,7 +755,7 @@ class InferenceRouter(Inference):
                             choices_data[idx] = {
                                 "content_parts": [],
                                 "tool_calls_builder": {},
-                                "finish_reason": None,
+                                "finish_reason": "stop",
                                 "logprobs_content_parts": [],
                             }
                         current_choice_data = choices_data[idx]

From 64d2306dd5481be3fdfaa11c415efe3f2788358e Mon Sep 17 00:00:00 2001
From: Derek Higgins <derekh@redhat.com>
Date: Thu, 4 Sep 2025 16:56:32 +0100
Subject: [PATCH 2/3] fix: distro-codegen pre-commit hook file pattern (#3337)

Update the file pattern from 'llama_stack/templates' to
'llama_stack/distributions' to properly trigger the Distribution
Template Codegen hook when distribution files change.

Signed-off-by: Derek Higgins <derekh@redhat.com>
---
 .pre-commit-config.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 514fe6d2e..b7880a9fc 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -86,7 +86,7 @@ repos:
         language: python
         pass_filenames: false
         require_serial: true
-        files: ^llama_stack/templates/.*$|^llama_stack/providers/.*/inference/.*/models\.py$
+        files: ^llama_stack/distributions/.*$|^llama_stack/providers/.*/inference/.*/models\.py$
       - id: provider-codegen
         name: Provider Codegen
         additional_dependencies:

From 85f33762d7cb7495b9aa5946284e4b6c85f5ad96 Mon Sep 17 00:00:00 2001
From: IAN MILLER <75687988+r3v5@users.noreply.github.com>
Date: Thu, 4 Sep 2025 17:15:13 +0100
Subject: [PATCH 3/3] refactor(server): remove hardcoded 409 and 404 status
 codes in server.py using httpx constants (#3333)

# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
This PR is eliminating hardcoded status codes: `409` CONFLICT and `404`
NOT_FOUND in `server.py` using `httpx` built-in constants. This
implementation will follow the existing structure to improve
readability, extensibility and developer experience. This is already was
implemented in #3131

<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->

## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
`./scripts/unit-tests.sh`
---
 llama_stack/core/server/server.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/llama_stack/core/server/server.py b/llama_stack/core/server/server.py
index d6dfc3435..b247a610d 100644
--- a/llama_stack/core/server/server.py
+++ b/llama_stack/core/server/server.py
@@ -132,9 +132,9 @@ def translate_exception(exc: Exception) -> HTTPException | RequestValidationErro
             },
         )
     elif isinstance(exc, ConflictError):
-        return HTTPException(status_code=409, detail=str(exc))
+        return HTTPException(status_code=httpx.codes.CONFLICT, detail=str(exc))
     elif isinstance(exc, ResourceNotFoundError):
-        return HTTPException(status_code=404, detail=str(exc))
+        return HTTPException(status_code=httpx.codes.NOT_FOUND, detail=str(exc))
     elif isinstance(exc, ValueError):
         return HTTPException(status_code=httpx.codes.BAD_REQUEST, detail=f"Invalid value: {str(exc)}")
     elif isinstance(exc, BadRequestError):