Describe the bug
A clear and concise description of what the bug is.
no guard when using the server should be easier to figure out
To Reproduce
from guardrails import settings
settings.use_remote = True
Guard(name="basic")(
model="gpt-4o-mini",
messages=[{
"role": "system",
"content": "Welcome to the OpenAI chatbot! How can I help you today?"
}]
)
should error if the guard isnt found
# Next, we created a Guarded endpoint that uses the hallucination validator:
guarded_client = OpenAI(
api_key=openai_api_key,
base_url="http://127.0.0.1:8000/guards/basic/openai/v1/"
)
guarded_client.chat.completions.create(
model="gpt-4o-mini",
messages=[{
"role": "system",
"content": "Welcome to the OpenAI chatbot! How can I help you today?"
}]
)
errors in an arcane server 500
Expected behavior
should be easier to figure out
Library version:
Version (0.5.X)
Additional context
Add any other context about the problem here.