"""
The ``mlflow.groq`` module provides an API for logging and loading Groq models.
"""
from mlflow.groq._groq_autolog import patched_call
from mlflow.utils.autologging_utils import autologging_integration, safe_patch
FLAVOR_NAME = "groq"
[docs]@autologging_integration(FLAVOR_NAME)
def autolog(
log_traces: bool = True,
disable: bool = False,
silent: bool = False,
):
"""
Enables (or disables) and configures autologging from Groq to MLflow.
Only synchronous calls are supported. Asynchronous APIs and streaming are not recorded.
Args:
log_traces: If ``True``, traces are logged for Groq models. If ``False``, no traces are
collected during inference. Default to ``True``.
disable: If ``True``, disables the Groq autologging. Default to ``False``.
silent: If ``True``, suppress all event logs and warnings from MLflow during Groq
autologging. If ``False``, show all events and warnings.
"""
from groq.resources.audio.transcriptions import Transcriptions
from groq.resources.audio.translations import Translations
from groq.resources.chat.completions import Completions as ChatCompletions
from groq.resources.embeddings import Embeddings
for task in (ChatCompletions, Translations, Transcriptions, Embeddings):
safe_patch(
FLAVOR_NAME,
task,
"create",
patched_call,
)