phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-20 03:40:05 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix: sambanova shields and model validation (#2693)

Browse source 
# What does this PR do?
Update the shield register validation of Sambanova not to raise, but
only warn when a model is not available in the base url endpoint used,
also added warnings when model is not available in the base url endpoint
used

<!-- 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.* -->
run starter distro with Sambanova enabled
This commit is contained in:
Jorge Piedrahita Ortiz 2025-07-11 15:29:15 -05:00 • committed by GitHub
parent 30b2e6a495
commit aa2595c7c3
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 31 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

21
llama_stack/providers/remote/safety/sambanova/sambanova.py
View file

@ -33,6 +33,7 @@ CANNED_RESPONSE_TEXT = "I can't answer that. Can I help with something else?"
class SambaNovaSafetyAdapter(Safety, ShieldsProtocolPrivate, NeedsRequestProviderData):
def __init__(self, config: SambaNovaSafetyConfig) -> None:
self.config = config
self.environment_available_models = []
async def initialize(self) -> None:
pass
@ -54,18 +55,18 @@ class SambaNovaSafetyAdapter(Safety, ShieldsProtocolPrivate, NeedsRequestProvide
async def register_shield(self, shield: Shield) -> None:
list_models_url = self.config.url + "/models"
try:
response = requests.get(list_models_url)
response.raise_for_status()
except requests.exceptions.RequestException as e:
raise RuntimeError(f"Request to {list_models_url} failed") from e
available_models = [model.get("id") for model in response.json().get("data", {})]
if len(self.environment_available_models) == 0:
try:
response = requests.get(list_models_url)
response.raise_for_status()
except requests.exceptions.RequestException as e:
raise RuntimeError(f"Request to {list_models_url} failed") from e
self.environment_available_models = [model.get("id") for model in response.json().get("data", {})]
if (
len(available_models) == 0
or "guard" not in shield.provider_resource_id.lower()
or shield.provider_resource_id.split("sambanova/")[-1] not in available_models
"guard" not in shield.provider_resource_id.lower()
or shield.provider_resource_id.split("sambanova/")[-1] not in self.environment_available_models
):
raise ValueError(f"Shield {shield.provider_resource_id} not found in SambaNova")
logger.warning(f"Shield {shield.provider_resource_id} not available in {list_models_url}")
async def run_shield(
self, shield_id: str, messages: list[Message], params: dict[str, Any] | None = None