openlit 1.34.19__py3-none-any.whl → 1.34.22__py3-none-any.whl

openlit/instrumentation/vertexai/vertexai.py

@@ -4,14 +4,15 @@ Module for monitoring VertexAI API calls.
 
 import logging
 import time
-from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
+from opentelemetry.trace import SpanKind
 from openlit.__helpers import (
-    get_chat_model_cost,
     handle_exception,
-    calculate_ttft,
-    calculate_tbt,
-    create_metrics_attributes,
+)
+from openlit.instrumentation.vertexai.utils import (
+    process_chunk,
+    process_chat_response,
+    process_streaming_chat_response,
+    extract_vertexai_details,
 )
 from openlit.semcov import SemanticConvention
 
@@ -21,35 +22,19 @@ logger = logging.getLogger(__name__)
 def send_message(version, environment, application_name, tracer,
              pricing_info, capture_message_content, metrics, disable_metrics):
     """
-    Generates a telemetry wrapper for messages to collect metrics.
-
-    Args:
-        version: Version of the monitoring package.
-        environment: Deployment environment (e.g., production, staging).
-        application_name: Name of the application using the VertexAI API.
-        tracer: OpenTelemetry tracer for creating spans.
-        pricing_info: Information used for calculating the cost of VertexAI usage.
-        capture_message_content: Flag indicating whether to trace the actual content.
-
-    Returns:
-        A function that wraps the chat method to add telemetry.
+    Generates a telemetry wrapper for VertexAI messages to collect metrics.
     """
 
     class TracedSyncStream:
         """
-        Wrapper for streaming responses to collect metrics and trace data.
-        Wraps the response to collect message IDs and aggregated response.
-
-        This class implements the '__aiter__' and '__anext__' methods that
-        handle asynchronous streaming responses.
-
-        This class also implements '__aenter__' and '__aexit__' methods that
-        handle asynchronous context management protocol.
+        Wrapper for streaming responses to collect telemetry.
         """
+
         def __init__(
                 self,
                 wrapped,
                 span,
+                span_name,
                 kwargs,
                 server_address,
                 server_port,
@@ -58,12 +43,10 @@ def send_message(version, environment, application_name, tracer,
             ):
             self.__wrapped__ = wrapped
             self._span = span
-            # Placeholder for aggregating streaming response
+            self._span_name = span_name
             self._llmresponse = ""
-            self._input_tokens = ""
-            self._output_tokens = ""
-
-            self._args = args
+            self._input_tokens = 0
+            self._output_tokens = 0
             self._kwargs = kwargs
             self._start_time = time.time()
             self._end_time = None
@@ -73,6 +56,7 @@ def send_message(version, environment, application_name, tracer,
             self._server_address = server_address
             self._server_port = server_port
             self._request_model = request_model
+            self._args = args
 
         def __enter__(self):
             self.__wrapped__.__enter__()
@@ -91,369 +75,67 @@ def send_message(version, environment, application_name, tracer,
         def __next__(self):
             try:
                 chunk = self.__wrapped__.__next__()
-                end_time = time.time()
-                # Record the timestamp for the current chunk
-                self._timestamps.append(end_time)
-
-                if len(self._timestamps) == 1:
-                    # Calculate time to first chunk
-                    self._ttft = calculate_ttft(self._timestamps, self._start_time)
-
-                self._llmresponse += str(chunk.text)
-                self._input_tokens = chunk.usage_metadata.prompt_token_count
-                self._output_tokens = chunk.usage_metadata.candidates_token_count
-
+                process_chunk(self, chunk)
                 return chunk
             except StopIteration:
-                # Handling exception ensure observability without disrupting operation
                 try:
-                    self._end_time = time.time()
-                    if len(self._timestamps) > 1:
-                        self._tbt = calculate_tbt(self._timestamps)
-
-                    # Format 'messages' into a single string
-                    message_prompt = self._kwargs.get("messages", "")
-                    formatted_messages = []
-                    for message in message_prompt:
-                        role = message["role"]
-                        content = message["content"]
-
-                        if isinstance(content, list):
-                            content_str_list = []
-                            for item in content:
-                                if item["type"] == "text":
-                                    content_str_list.append(f'text: {item["text"]}')
-                                elif (item["type"] == "image_url" and
-                                      not item["image_url"]["url"].startswith("data:")):
-                                    content_str_list.append(f'image_url: {item["image_url"]["url"]}')
-                            content_str = ", ".join(content_str_list)
-                            formatted_messages.append(f"{role}: {content_str}")
-                        else:
-                            formatted_messages.append(f"{role}: {content}")
-                    prompt = "\n".join(formatted_messages) or str(self._args[0][0])
-
-                    # Calculate cost of the operation
-                    cost = get_chat_model_cost(self._request_model,
-                                                pricing_info, self._input_tokens,
-                                                self._output_tokens)
-
-                    # Set Span attributes (OTel Semconv)
-                    self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                    self._span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_SYSTEM,
-                                        SemanticConvention.GEN_AI_SYSTEM_VERTEXAI)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                        self._request_model)
-                    self._span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        self._server_port)
-
-                    inference_config = self._kwargs.get('generation_config', {})
-
-                    # List of attributes and their config keys
-                    attributes = [
-                        (SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY, 'frequency_penalty'),
-                        (SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, 'max_output_tokens'),
-                        (SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY, 'presence_penalty'),
-                        (SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES, 'stop_sequences'),
-                        (SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, 'temperature'),
-                        (SemanticConvention.GEN_AI_REQUEST_TOP_P, 'top_p'),
-                        (SemanticConvention.GEN_AI_REQUEST_TOP_K, 'top_k'),
-                    ]
-
-                    # Set each attribute if the corresponding value exists and is not None
-                    for attribute, key in attributes:
-                        # Use the `get` method to safely access keys in the dictionary
-                        value = inference_config.get(key)
-                        if value is not None:
-                            self._span.set_attribute(attribute, value)
-
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                        self._request_model)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
-                                        self._input_tokens)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
-                                        self._output_tokens)
-                    self._span.set_attribute(SemanticConvention.SERVER_ADDRESS,
-                                        self._server_address)
-                    if isinstance(self._llmresponse, str):
-                        self._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                        "text")
-                    else:
-                        self._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                        "json")
-
-                    # Set Span attributes (Extra)
-                    self._span.set_attribute(DEPLOYMENT_ENVIRONMENT,
-                                        environment)
-                    self._span.set_attribute(SERVICE_NAME,
-                                        application_name)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IS_STREAM,
-                                        True)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_TOTAL_TOKENS,
-                                        self._input_tokens + self._output_tokens)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                        cost)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TBT,
-                                        self._tbt)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TTFT,
-                                        self._ttft)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION,
-                                        version)
-                    if capture_message_content:
-                        self._span.add_event(
-                            name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
-                            attributes={
-                                SemanticConvention.GEN_AI_CONTENT_PROMPT: prompt,
-                            },
-                        )
-                        self._span.add_event(
-                            name=SemanticConvention.GEN_AI_CONTENT_COMPLETION_EVENT,
-                            attributes={
-                                SemanticConvention.GEN_AI_CONTENT_COMPLETION: self._llmresponse,
-                            },
-                        )
-                    self._span.set_status(Status(StatusCode.OK))
-
-                    if disable_metrics is False:
-                        attributes = create_metrics_attributes(
-                            service_name=application_name,
-                            deployment_environment=environment,
-                            operation=SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
-                            system=SemanticConvention.GEN_AI_SYSTEM_VERTEXAI,
-                            request_model=self._request_model,
-                            server_address=self._server_address,
-                            server_port=self._server_port,
-                            response_model=self._request_model,
-                        )
-
-                        metrics["genai_client_usage_tokens"].record(
-                            self._input_tokens + self._output_tokens, attributes
-                        )
-                        metrics["genai_client_operation_duration"].record(
-                            self._end_time - self._start_time, attributes
-                        )
-                        metrics["genai_server_tbt"].record(
-                            self._tbt, attributes
-                        )
-                        metrics["genai_server_ttft"].record(
-                            self._ttft, attributes
+                    with self._span:
+                        process_streaming_chat_response(
+                            self,
+                            pricing_info=pricing_info,
+                            environment=environment,
+                            application_name=application_name,
+                            metrics=metrics,
+                            capture_message_content=capture_message_content,
+                            disable_metrics=disable_metrics,
+                            version=version
                         )
-                        metrics["genai_requests"].add(1, attributes)
-                        metrics["genai_completion_tokens"].add(self._output_tokens, attributes)
-                        metrics["genai_prompt_tokens"].add(self._input_tokens, attributes)
-                        metrics["genai_cost"].record(cost, attributes)
-
                 except Exception as e:
                     handle_exception(self._span, e)
-                    logger.error("Error in trace creation: %s", e)
-                finally:
-                    self._span.end()
                 raise
 
     def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the 'messages' API call to add telemetry.
-
-        This collects metrics such as execution time, cost, and token usage, and handles errors
-        gracefully, adding details to the trace for observability.
-
-        Args:
-            wrapped: The original 'messages' method to be wrapped.
-            instance: The instance of the class where the original method is defined.
-            args: Positional arguments for the 'messages' method.
-            kwargs: Keyword arguments for the 'messages' method.
-
-        Returns:
-            The response from the original 'messages' method.
+        Wraps the VertexAI API call to add telemetry.
         """
 
-        # Check if streaming is enabled for the API call
         streaming = kwargs.get("stream", False)
-
-        try:
-            location = instance._model._location
-            request_model = "/".join(instance._model._model_name.split("/")[3:])
-        except:
-            location = instance._location
-            request_model = "/".join(instance._model_name.split("/")[3:])
-
-        server_address, server_port = location + '-aiplatform.googleapis.com', 443
+        server_address, server_port, request_model = extract_vertexai_details(instance)
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
 
-        # pylint: disable=no-else-return
         if streaming:
-            # Special handling for streaming response to accommodate the nature of data flow
             awaited_wrapped = wrapped(*args, **kwargs)
             span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
 
-            return TracedSyncStream(awaited_wrapped, span, kwargs, server_address, server_port, request_model, args)
+            return TracedSyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port, request_model, args)
 
-        # Handling for non-streaming responses
         else:
             with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
                 start_time = time.time()
                 response = wrapped(*args, **kwargs)
-                end_time = time.time()
 
                 try:
-                    # Format 'messages' into a single string
-                    message_prompt = kwargs.get("contents", [])
-                    formatted_messages = []
-
-                    for content in message_prompt:
-                        role = content.role
-                        parts = content.parts
-                        content_str = []
-
-                        for part in parts:
-                            # Collect relevant fields and handle each type of data that Part could contain
-                            if part.text:
-                                content_str.append(f"text: {part.text}")
-                            if part.video_metadata:
-                                content_str.append(f"video_metadata: {part.video_metadata}")
-                            if part.thought:
-                                content_str.append(f"thought: {part.thought}")
-                            if part.code_execution_result:
-                                content_str.append(f"code_execution_result: {part.code_execution_result}")
-                            if part.executable_code:
-                                content_str.append(f"executable_code: {part.executable_code}")
-                            if part.file_data:
-                                content_str.append(f"file_data: {part.file_data}")
-                            if part.function_call:
-                                content_str.append(f"function_call: {part.function_call}")
-                            if part.function_response:
-                                content_str.append(f"function_response: {part.function_response}")
-                            if part.inline_data:
-                                content_str.append(f"inline_data: {part.inline_data}")
-
-                        formatted_messages.append(f"{role}: {', '.join(content_str)}")
-
-                    prompt = "\n".join(formatted_messages) or str(args[0][0])
-
-                    input_tokens = response.usage_metadata.prompt_token_count
-                    output_tokens = response.usage_metadata.candidates_token_count
-
-                    # Calculate cost of the operation
-                    cost = get_chat_model_cost(request_model,
-                                                pricing_info, input_tokens,
-                                                output_tokens)
-
-                    # Set base span attribues (OTel Semconv)
-                    span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                    span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT)
-                    span.set_attribute(SemanticConvention.GEN_AI_SYSTEM,
-                                        SemanticConvention.GEN_AI_SYSTEM_VERTEXAI)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                        request_model)
-                    span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        server_port)
-
-                    inference_config = kwargs.get('generation_config', {})
-
-                    # List of attributes and their config keys
-                    attributes = [
-                        (SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY, 'frequency_penalty'),
-                        (SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, 'max_output_tokens'),
-                        (SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY, 'presence_penalty'),
-                        (SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES, 'stop_sequences'),
-                        (SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, 'temperature'),
-                        (SemanticConvention.GEN_AI_REQUEST_TOP_P, 'top_p'),
-                        (SemanticConvention.GEN_AI_REQUEST_TOP_K, 'top_k'),
-                    ]
-
-                    # Set each attribute if the corresponding value exists and is not None
-                    for attribute, key in attributes:
-                        # Use the `get` method to safely access keys in the dictionary
-                        value = inference_config.get(key)
-                        if value is not None:
-                            span.set_attribute(attribute, value)
-
-                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                        request_model)
-                    span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
-                                        input_tokens)
-                    span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
-                                        output_tokens)
-                    span.set_attribute(SemanticConvention.SERVER_ADDRESS,
-                                        server_address)
-                    # span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON,
-                    #                     [str(response.candidates[0].finish_reason)])
-
-                    # Set base span attribues (Extras)
-                    span.set_attribute(DEPLOYMENT_ENVIRONMENT,
-                                        environment)
-                    span.set_attribute(SERVICE_NAME,
-                                        application_name)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IS_STREAM,
-                                        False)
-                    span.set_attribute(SemanticConvention.GEN_AI_USAGE_TOTAL_TOKENS,
-                                        input_tokens + output_tokens)
-                    span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                        cost)
-                    span.set_attribute(SemanticConvention.GEN_AI_SERVER_TTFT,
-                                        end_time - start_time)
-                    span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION,
-                                        version)
-                    if capture_message_content:
-                        span.add_event(
-                            name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
-                            attributes={
-                                SemanticConvention.GEN_AI_CONTENT_PROMPT: prompt,
-                            },
-                        )
-                        span.add_event(
-                            name=SemanticConvention.GEN_AI_CONTENT_COMPLETION_EVENT,
-                            attributes={
-                                SemanticConvention.GEN_AI_CONTENT_COMPLETION: response.text,
-                            },
-                        )
-
-                        if isinstance(response.text, str):
-                            span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                            "text")
-                        elif response.text is not None:
-                            span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                            "json")
-
-                    span.set_status(Status(StatusCode.OK))
-
-                    if disable_metrics is False:
-                        attributes = create_metrics_attributes(
-                            service_name=application_name,
-                            deployment_environment=environment,
-                            operation=SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
-                            system=SemanticConvention.GEN_AI_SYSTEM_VERTEXAI,
-                            request_model=request_model,
-                            server_address=server_address,
-                            server_port=server_port,
-                            response_model=request_model,
-                        )
-
-                        metrics["genai_client_usage_tokens"].record(
-                            input_tokens + output_tokens, attributes
-                        )
-                        metrics["genai_client_operation_duration"].record(
-                            end_time - start_time, attributes
-                        )
-                        metrics["genai_server_ttft"].record(
-                            end_time - start_time, attributes
-                        )
-                        metrics["genai_requests"].add(1, attributes)
-                        metrics["genai_completion_tokens"].add(output_tokens, attributes)
-                        metrics["genai_prompt_tokens"].add(input_tokens, attributes)
-                        metrics["genai_cost"].record(cost, attributes)
-
-                    # Return original response
-                    return response
+                    response = process_chat_response(
+                        response=response,
+                        request_model=request_model,
+                        pricing_info=pricing_info,
+                        server_port=server_port,
+                        server_address=server_address,
+                        environment=environment,
+                        application_name=application_name,
+                        metrics=metrics,
+                        start_time=start_time,
+                        span=span,
+                        capture_message_content=capture_message_content,
+                        disable_metrics=disable_metrics,
+                        version=version,
+                        **kwargs
+                    )
 
                 except Exception as e:
                     handle_exception(span, e)
-                    logger.error("Error in trace creation: %s", e)
 
-                    # Return original response
-                    return response
+                return response
 
     return wrapper

{openlit-1.34.19.dist-info → openlit-1.34.22.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: openlit
-Version: 1.34.19
+Version: 1.34.22
 Summary: OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects
 License: Apache-2.0
 Keywords: OpenTelemetry,otel,otlp,llm,tracing,openai,anthropic,claude,cohere,llm monitoring,observability,monitoring,gpt,Generative AI,chatGPT,gpu

{openlit-1.34.19.dist-info → openlit-1.34.22.dist-info}/RECORD

@@ -1,4 +1,4 @@
-openlit/__helpers.py,sha256=x_HA-B3v0lawXeg3_yASXAzN0P0hChrgWyYXdLGY0Pw,14862
+openlit/__helpers.py,sha256=UmW0E97r7UUps2rkdQ-S_lXhaeH3Txp-umyAOon45mk,16458
 openlit/__init__.py,sha256=ris6-GY0ePSbK_jvawHTXymGClVF7yeKdIT95IRBl18,24086
 openlit/evals/__init__.py,sha256=nJe99nuLo1b5rf7pt9U9BCdSDedzbVi2Fj96cgl7msM,380
 openlit/evals/all.py,sha256=oWrue3PotE-rB5WePG3MRYSA-ro6WivkclSHjYlAqGs,7154
@@ -33,9 +33,9 @@ openlit/instrumentation/azure_ai_inference/__init__.py,sha256=_GuYy4ypF6_HICpAC8
 openlit/instrumentation/azure_ai_inference/async_azure_ai_inference.py,sha256=SFrniRWPqVBxFJVOpC8w1qNGSYZhCFXeKHVHu5pEdZI,5906
 openlit/instrumentation/azure_ai_inference/azure_ai_inference.py,sha256=hRFuuvaXflctNNbk7N2GOfKaC_eCHbrBWf9_1sZcaGY,5808
 openlit/instrumentation/azure_ai_inference/utils.py,sha256=JqKZgb6VppDbAQ2RdH_dQ2oWVnaqGIA1PCmMl1yrMtA,15491
-openlit/instrumentation/bedrock/__init__.py,sha256=Sfd0vm4Dfm1t-N7vBPRwU57GLTlZP2M4rVYRek_JHXY,1625
-openlit/instrumentation/bedrock/bedrock.py,sha256=kP9ESKzqhWu-dIWseyaeyendUyo6b7xJwjGo3LGi5Jc,2817
-openlit/instrumentation/bedrock/utils.py,sha256=_mTUIbioEg4jfoxocUbfc7RgGjhm9ACelbxIoFu4jbM,11636
+openlit/instrumentation/bedrock/__init__.py,sha256=D4Wyjb9kJASUQMy5hI41Mvp_csP4k4F6tASdQjt8IZ0,1681
+openlit/instrumentation/bedrock/bedrock.py,sha256=_2oTa9cf-OxHTFCmNUWT__vHnv7-wUcxm9-6nC1K3jg,8036
+openlit/instrumentation/bedrock/utils.py,sha256=7qn8pI-BQcW8ZLxYJb_W3Yzncqbz4QBCR9f2u9N7C4Y,9345
 openlit/instrumentation/chroma/__init__.py,sha256=4ZeHY1OInRKQbb4qg8BVvGJtWN1XdzW6mosqi7-6ruE,3353
 openlit/instrumentation/chroma/chroma.py,sha256=fPIX6vgcDFkBgKGxrkXZWDaBtZa5y-PV3jJjKsFqjDg,10640
 openlit/instrumentation/cohere/__init__.py,sha256=FIJ_QokUZcN9UsVPWckRfHiLVJ5zj3jtcOjQmjQA5f0,2978
@@ -82,7 +82,7 @@ openlit/instrumentation/langchain/langchain.py,sha256=zgfzfOIDsaRoVgWl1T4XX2CLO7
 openlit/instrumentation/letta/__init__.py,sha256=K8PtRKxuueyqEYE3LzxWJ74IieNKSI6dmk9sNRd8Mt0,3031
 openlit/instrumentation/letta/letta.py,sha256=SCIpJ4tdB1l1BmeQx4raaTS4MQO5X15pLvS4PepEKBE,8481
 openlit/instrumentation/litellm/__init__.py,sha256=D47yfDLLEKpkaRAy7_Yif70kj88AGqLQYZAABpTN4sE,2284
-openlit/instrumentation/litellm/async_litellm.py,sha256=6cL_hv9t4tuXkcKZvpTdnb0wGTs54lSwGWCtdYZvyXg,6768
+openlit/instrumentation/litellm/async_litellm.py,sha256=GdMXyCLLuC9-7nXTsKTgvXOOQnw0oTbUwL_DYp_pNoE,6780
 openlit/instrumentation/litellm/litellm.py,sha256=xLna3I_jcywTtIs1tBjHAQKyKjNM07T8GHX9pIqZcQ0,6664
 openlit/instrumentation/litellm/utils.py,sha256=VMSnYkKn9yZtOphIh2ENNuqJtGjz1fXEYUKi5JGHC7A,13195
 openlit/instrumentation/llamaindex/__init__.py,sha256=2pmd9BKw3ab0OJ4yuJEg0-Jkn_haDbXvbUm5r2-rOCU,2007
@@ -102,9 +102,10 @@ openlit/instrumentation/ollama/__init__.py,sha256=WxjqjuR8ovMU5dR08OELNqClbuM7ns
 openlit/instrumentation/ollama/async_ollama.py,sha256=ORXwem8lgSrhOcci55NkChIK9SNc3IYIpLjF_ogsGA8,6666
 openlit/instrumentation/ollama/ollama.py,sha256=8mvrWfU1c5h1L7lxWo47YBJ7g2u7QZmSZuuP0URtTDo,6538
 openlit/instrumentation/ollama/utils.py,sha256=TIE3_ur2U-iyCclna7TzwjDIFC9PZjRnZqNDV6NfG-0,11958
-openlit/instrumentation/openai/__init__.py,sha256=KI3ncllea3VzK0lvBfZXBhs2EClSLe38WEPdIL4_SOo,6311
-openlit/instrumentation/openai/async_openai.py,sha256=JyA8MDxWCM38Te6mJzBdfonRgIIlo2ziLn7HOmzqxxo,81398
-openlit/instrumentation/openai/openai.py,sha256=5fgRyK5dUN2zUdrN0vBSZFnSEAXf2dKS0qnq_85-mQE,81175
+openlit/instrumentation/openai/__init__.py,sha256=FKdqQLidU62qSdVL3ADK-_9JzwHE5Y8UKncVVP1KWXs,5594
+openlit/instrumentation/openai/async_openai.py,sha256=QvEEKZnZYl9Vf-wsX1voTMMZed1eNhRI9aUT8CtFJi0,18003
+openlit/instrumentation/openai/openai.py,sha256=34_FqOwSroNOm_mmLzZb8Y7xtr5StwnUyRQmHP6HHJc,17698
+openlit/instrumentation/openai/utils.py,sha256=ZblnWf9gN9KsHOqa4nZ6AsxOhBHVteHpHJXvcPR9b44,35365
 openlit/instrumentation/openai_agents/__init__.py,sha256=tRTSIrUtkXc_lfQnVanXmQLd2Sy9RqBNTHF5FhhZx7o,1530
 openlit/instrumentation/openai_agents/openai_agents.py,sha256=kRWPgjofcOviMi3w7CsRvJO3SCjqPmuq-PM800vIM7g,2678
 openlit/instrumentation/phidata/__init__.py,sha256=tqls5-UI6FzbjxYgq_qqAfALhWJm8dHn2NtgqiQA4f8,1557
@@ -131,9 +132,10 @@ openlit/instrumentation/together/utils.py,sha256=n7r_pM_sqFnJEAkL7OhPydr0Uct0A74
 openlit/instrumentation/transformers/__init__.py,sha256=hXq0WUZNl6Sz0Ihk29kA9i8Q1j0e1URFb7v7etnQpxI,1511
 openlit/instrumentation/transformers/transformers.py,sha256=MHnHVo_6NP0gSIqxen6qQpCrZ0fs8Ec80EdZumMpVNo,1797
 openlit/instrumentation/transformers/utils.py,sha256=MMy_SyRyDI4X-0mqbBwStac0xabmw0ZRvv_VWLA_Nkg,8426
-openlit/instrumentation/vertexai/__init__.py,sha256=mT28WCBvQfRCkAWGL6bd0EjEPHvMjaNcz6T3jsLZh8k,3745
-openlit/instrumentation/vertexai/async_vertexai.py,sha256=-kpg-eiL76O5_XopUPghCYwJHf0Nrxi00_Z5tCwq6zM,23086
-openlit/instrumentation/vertexai/vertexai.py,sha256=5NB090aWlm9DnlccNNLRO6A97P_RN-JnHb5JS01tYyw,23000
+openlit/instrumentation/vertexai/__init__.py,sha256=ti0LBtnXVN8cC8AFkn_52JSKamfR5AL9MdNac9u1R28,3640
+openlit/instrumentation/vertexai/async_vertexai.py,sha256=CxogAXam2ZUpFaIyWI3llaJn5bXAtKYm0L3dYEWu278,4865
+openlit/instrumentation/vertexai/utils.py,sha256=7B5bYcBNLNpt3uG64yG0UUMMyfaqSRVPDaeebppBMmk,8425
+openlit/instrumentation/vertexai/vertexai.py,sha256=tHcnpw4s0ugaxP4hG5q3BFnUtLBcoEuqARdwR-dmf3c,4775
 openlit/instrumentation/vllm/__init__.py,sha256=uaSzQmgDuKJ-sh61sfVdzVt2qAZaozZIQ8sbmQ0XpZE,1357
 openlit/instrumentation/vllm/utils.py,sha256=HuCPNBgChWg9vA7DHNFCij_y8qj27DjZxdZ0Nvdt2fg,5751
 openlit/instrumentation/vllm/vllm.py,sha256=VzazF2f4LLwjZDO_G8lIN_d622oSJM0fIO9wjxXbhyg,2004
@@ -141,7 +143,7 @@ openlit/otel/events.py,sha256=VrMjTpvnLtYRBHCiFwJojTQqqNpRCxoD4yJYeQrtPsk,3560
 openlit/otel/metrics.py,sha256=GM2PDloBGRhBTkHHkYaqmOwIAQkY124ZhW4sEqW1Fgk,7086
 openlit/otel/tracing.py,sha256=tjV2bEbEDPUB1Z46gE-UsJsb04sRdFrfbhIDkxViZc0,3103
 openlit/semcov/__init__.py,sha256=8oIh2VC667NDh8FA3M-ESusHmeus1sgDUD8binx_nAc,13519
-openlit-1.34.19.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-openlit-1.34.19.dist-info/METADATA,sha256=dxAiXXKjEgxrLrnGShuRAbVebbg1GOmPaa5CODpVqCA,23470
-openlit-1.34.19.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-openlit-1.34.19.dist-info/RECORD,,
+openlit-1.34.22.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+openlit-1.34.22.dist-info/METADATA,sha256=ADqRel4c0tZ3_W6h7dUSHh93x56YyU8sHYdjwQZlJVg,23470
+openlit-1.34.22.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+openlit-1.34.22.dist-info/RECORD,,