openlit 1.34.1__tar.gz → 1.34.3__tar.gz

{openlit-1.34.1 → openlit-1.34.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: openlit
-Version: 1.34.1
+Version: 1.34.3
 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-1.34.1 → openlit-1.34.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "openlit"
-version = "1.34.1"
+version = "1.34.3"
 description = "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects"
 authors = ["OpenLIT"]
 license = "Apache-2.0"

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/google_ai_studio/async_google_ai_studio.py

@@ -59,14 +59,12 @@ def async_generate(version, environment, application_name,
                         version=version,
                 )
 
-                return response
-
             except Exception as e:
                 handle_exception(span, e)
                 logger.error("Error in trace creation: %s", e)
 
-                # Return original response
-                return response
+            # Return original response
+            return response
 
     return wrapper
 

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/google_ai_studio/google_ai_studio.py

@@ -59,14 +59,12 @@ def generate(version, environment, application_name,
                         version=version,
                 )
 
-                return response
-
             except Exception as e:
                 handle_exception(span, e)
                 logger.error("Error in trace creation: %s", e)
 
-                # Return original response
-                return response
+            # Return original response
+            return response
 
     return wrapper
 

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/google_ai_studio/utils.py

@@ -2,10 +2,8 @@
 Google AI Studio OpenTelemetry instrumentation utility functions
 """
 import time
-
 from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
 from opentelemetry.trace import Status, StatusCode
-
 from openlit.__helpers import (
     calculate_ttft,
     response_as_dict,
@@ -117,6 +115,7 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
     scope._span.set_attribute(SemanticConvention.GEN_AI_SYSTEM, SemanticConvention.GEN_AI_SYSTEM_GEMINI)
     scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL, request_model)
     scope._span.set_attribute(SemanticConvention.SERVER_PORT, scope._server_port)
+    scope._span.set_attribute(SemanticConvention.SERVER_ADDRESS, scope._server_address)
 
     inference_config = scope._kwargs.get('config', {})
 
@@ -142,7 +141,6 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, scope._input_tokens)
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, scope._output_tokens)
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_REASONING_TOKENS, scope._reasoning_tokens)
-    scope._span.set_attribute(SemanticConvention.SERVER_ADDRESS, scope._server_address)
 
     scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
                               'text' if isinstance(scope._llmresponse, str) else 'json')

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/openai/__init__.py

@@ -5,9 +5,10 @@ import importlib.metadata
 from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
 from wrapt import wrap_function_wrapper
 
-from openlit.instrumentation.openai.openai import chat_completions, embedding, responses
+from openlit.instrumentation.openai.openai import chat_completions, embedding, responses, chat_completions_parse
 from openlit.instrumentation.openai.openai import image_generate, image_variatons, audio_create
-from openlit.instrumentation.openai.async_openai import async_chat_completions, async_embedding
+from openlit.instrumentation.openai.async_openai import (async_chat_completions, async_embedding,
+                                                         async_chat_completions_parse)
 from openlit.instrumentation.openai.async_openai import async_image_generate, async_image_variatons
 from openlit.instrumentation.openai.async_openai import async_audio_create, async_responses
 
@@ -127,6 +128,20 @@ class OpenAIInstrumentor(BaseInstrumentor):
                                metrics, disable_metrics),
         )
 
+        wrap_function_wrapper(
+            "openai.resources.beta.chat.completions",
+            "Completions.parse",
+            chat_completions_parse(version, environment, application_name, tracer, pricing_info,
+                                   capture_message_content, metrics, disable_metrics),
+        )
+
+        wrap_function_wrapper(
+            "openai.resources.beta.chat.completions",
+            "AsyncCompletions.parse",
+            async_chat_completions_parse(version, environment, application_name, tracer, pricing_info,
+                                         capture_message_content, metrics, disable_metrics),
+        )
+
     @staticmethod
     def _uninstrument(self, **kwargs):
         pass

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/openai/async_openai.py

@@ -882,6 +882,167 @@ def async_chat_completions(version, environment, application_name,
 
     return wrapper
 
+def async_chat_completions_parse(version, environment, application_name, tracer, pricing_info, capture_message_content,
+                                 metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for chat completions parse 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 OpenAI API.
+        tracer: OpenTelemetry tracer for creating spans.
+        pricing_info: Information used for calculating the cost of OpenAI usage.
+        capture_message_content: Flag indicating whether to trace the actual content.
+
+    Returns:
+        A function that wraps the chat completions parse method to add telemetry.
+    """
+
+    async def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the 'chat.completions.parse' 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 'chat.completions' method to be wrapped.
+            instance: The instance of the class where the original method is defined.
+            args: Positional arguments for the 'chat.completions' method.
+            kwargs: Keyword arguments for the 'chat.completions' method.
+
+        Returns:
+            The response from the original 'chat.completions.parse' method.
+        """
+        server_address, server_port = set_server_address_and_port(instance, "api.openai.com", 443)
+        request_model = kwargs.get("model", "gpt-4o")
+        span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+
+        with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+            start_time = time.time()
+            try:
+                # Execute the original 'parse' method
+                response = await wrapped(*args, **kwargs)
+                end_time = time.time()
+
+                response_dict = response_as_dict(response)
+
+                # Format 'messages' from kwargs to calculate input tokens
+                message_prompt = kwargs.get("messages", "")
+                formatted_messages = []
+                for message in message_prompt:
+                    role = message.get("role")
+                    content = message.get("content")
+                    if content:
+                        formatted_messages.append(f"{role}: {content}")
+                prompt = "\n".join(formatted_messages)
+
+                input_tokens = response_dict.get('usage').get('prompt_tokens')
+                output_tokens = response_dict.get('usage').get('completion_tokens')
+
+                # Calculate cost
+                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_OPENAI)
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL, request_model)
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_SEED, str(kwargs.get("seed", "")))
+                span.set_attribute(SemanticConvention.SERVER_PORT, server_port)
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+                                   str(kwargs.get("frequency_penalty", 0.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, str(kwargs.get("max_tokens", -1)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY,
+                                   str(kwargs.get("presence_penalty", 0.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES, str(kwargs.get("stop", [])))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, str(kwargs.get("temperature", 1.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P, str(kwargs.get("top_p", 1.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID, response_dict.get("id"))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL, response_dict.get('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_REQUEST_SERVICE_TIER,
+                                   str(kwargs.get("service_tier", "auto")))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_SERVICE_TIER,
+                                   response_dict.get('service_tier', 'auto'))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_SYSTEM_FINGERPRINT,
+                                   str(response_dict.get('system_fingerprint', '')))
+
+                # 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_USER, kwargs.get("user", ""))
+                span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION, version)
+                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)
+
+                if capture_message_content:
+                    span.add_event(
+                        name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
+                        attributes={SemanticConvention.GEN_AI_CONTENT_PROMPT: prompt},
+                    )
+
+                for i in range(kwargs.get('n', 1)):
+                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON,
+                                       [response_dict.get('choices')[i].get('finish_reason')])
+                    if capture_message_content:
+                        span.add_event(
+                            name=SemanticConvention.GEN_AI_CONTENT_COMPLETION_EVENT,
+                            attributes={
+                                # pylint: disable=line-too-long
+                                SemanticConvention.GEN_AI_CONTENT_COMPLETION: str(
+                                    response_dict.get('choices')[i].get('message').get('content')),
+                            },
+                        )
+                    if kwargs.get('tools'):
+                        span.set_attribute(SemanticConvention.GEN_AI_TOOL_CALLS,
+                                           str(response_dict.get('choices')[i].get('message').get('tool_calls')))
+
+                    if isinstance(response_dict.get('choices')[i].get('message').get('content'), str):
+                        span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
+                                           "text")
+                    elif response_dict.get('choices')[i].get('message').get('content') is not None:
+                        span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
+                                           "json")
+
+                span.set_status(Status(StatusCode.OK))
+
+                if not disable_metrics:
+                    attributes = create_metrics_attributes(
+                        service_name=application_name,
+                        deployment_environment=environment,
+                        operation=SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+                        system=SemanticConvention.GEN_AI_SYSTEM_OPENAI,
+                        request_model=request_model,
+                        server_address=server_address,
+                        server_port=server_port,
+                        response_model=response_dict.get('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 response
+
+            except Exception as e:
+                handle_exception(span, e)
+                logger.error("Error in 'parse' trace creation: %s", e)
+                # Re-raise the exception to not interfere with the application flow
+                raise
+
+    return wrapper
+
 def async_embedding(version, environment, application_name,
               tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/openai/openai.py

@@ -882,6 +882,167 @@ def chat_completions(version, environment, application_name,
 
     return wrapper
 
+def chat_completions_parse(version, environment, application_name, tracer, pricing_info, capture_message_content,
+                           metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for chat completions parse 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 OpenAI API.
+        tracer: OpenTelemetry tracer for creating spans.
+        pricing_info: Information used for calculating the cost of OpenAI usage.
+        capture_message_content: Flag indicating whether to trace the actual content.
+
+    Returns:
+        A function that wraps the chat completions parse method to add telemetry.
+    """
+
+    def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the 'chat.completions.parse' 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 'chat.completions' method to be wrapped.
+            instance: The instance of the class where the original method is defined.
+            args: Positional arguments for the 'chat.completions' method.
+            kwargs: Keyword arguments for the 'chat.completions' method.
+
+        Returns:
+            The response from the original 'chat.completions.parse' method.
+        """
+        server_address, server_port = set_server_address_and_port(instance, "api.openai.com", 443)
+        request_model = kwargs.get("model", "gpt-4o")
+        span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+
+        with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+            start_time = time.time()
+            try:
+                # Execute the original 'parse' method
+                response = wrapped(*args, **kwargs)
+                end_time = time.time()
+
+                response_dict = response_as_dict(response)
+
+                # Format 'messages' from kwargs to calculate input tokens
+                message_prompt = kwargs.get("messages", "")
+                formatted_messages = []
+                for message in message_prompt:
+                    role = message.get("role")
+                    content = message.get("content")
+                    if content:
+                        formatted_messages.append(f"{role}: {content}")
+                prompt = "\n".join(formatted_messages)
+
+                input_tokens = response_dict.get('usage').get('prompt_tokens')
+                output_tokens = response_dict.get('usage').get('completion_tokens')
+
+                # Calculate cost
+                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_OPENAI)
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL, request_model)
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_SEED, str(kwargs.get("seed", "")))
+                span.set_attribute(SemanticConvention.SERVER_PORT, server_port)
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+                                   str(kwargs.get("frequency_penalty", 0.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, str(kwargs.get("max_tokens", -1)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY,
+                                   str(kwargs.get("presence_penalty", 0.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES, str(kwargs.get("stop", [])))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, str(kwargs.get("temperature", 1.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P, str(kwargs.get("top_p", 1.0)))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID, response_dict.get("id"))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL, response_dict.get('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_REQUEST_SERVICE_TIER,
+                                   str(kwargs.get("service_tier", "auto")))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_SERVICE_TIER,
+                                   response_dict.get('service_tier', 'auto'))
+                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_SYSTEM_FINGERPRINT,
+                                   str(response_dict.get('system_fingerprint', '')))
+
+                # 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_USER, kwargs.get("user", ""))
+                span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION, version)
+                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)
+
+                if capture_message_content:
+                    span.add_event(
+                        name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
+                        attributes={SemanticConvention.GEN_AI_CONTENT_PROMPT: prompt},
+                    )
+
+                for i in range(kwargs.get('n', 1)):
+                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON,
+                                       [response_dict.get('choices')[i].get('finish_reason')])
+                    if capture_message_content:
+                        span.add_event(
+                            name=SemanticConvention.GEN_AI_CONTENT_COMPLETION_EVENT,
+                            attributes={
+                                # pylint: disable=line-too-long
+                                SemanticConvention.GEN_AI_CONTENT_COMPLETION: str(
+                                    response_dict.get('choices')[i].get('message').get('content')),
+                            },
+                        )
+                    if kwargs.get('tools'):
+                        span.set_attribute(SemanticConvention.GEN_AI_TOOL_CALLS,
+                                           str(response_dict.get('choices')[i].get('message').get('tool_calls')))
+
+                    if isinstance(response_dict.get('choices')[i].get('message').get('content'), str):
+                        span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
+                                           "text")
+                    elif response_dict.get('choices')[i].get('message').get('content') is not None:
+                        span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
+                                           "json")
+
+                span.set_status(Status(StatusCode.OK))
+
+                if not disable_metrics:
+                    attributes = create_metrics_attributes(
+                        service_name=application_name,
+                        deployment_environment=environment,
+                        operation=SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+                        system=SemanticConvention.GEN_AI_SYSTEM_OPENAI,
+                        request_model=request_model,
+                        server_address=server_address,
+                        server_port=server_port,
+                        response_model=response_dict.get('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 response
+
+            except Exception as e:
+                handle_exception(span, e)
+                logger.error("Error in 'parse' trace creation: %s", e)
+                # Re-raise the exception to not interfere with the application flow
+                raise
+
+    return wrapper
+
 def embedding(version, environment, application_name,
               tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """

{openlit-1.34.1 → openlit-1.34.3}/src/openlit/instrumentation/vllm/__init__.py

@@ -32,7 +32,7 @@ class VLLMInstrumentor(BaseInstrumentor):
 
         # sync chat
         wrap_function_wrapper(
-            "vllm",
+            "vllm.entrypoints.llm",
             "LLM.generate",
             generate(version, environment, application_name,
                   tracer, pricing_info, capture_message_content, metrics, disable_metrics),

openlit-1.34.3/src/openlit/instrumentation/vllm/utils.py

@@ -0,0 +1,161 @@
+"""
+Utility functions for vLLM instrumentation.
+"""
+
+import time
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
+from opentelemetry.trace import Status, StatusCode
+from openlit.__helpers import (
+    calculate_tbt,
+    get_chat_model_cost,
+    general_tokens,
+    create_metrics_attributes,
+)
+from openlit.semcov import SemanticConvention
+
+def get_inference_config(args, kwargs):
+    """
+    Safely extract inference configuration from args or kwargs.
+    """
+
+    if 'sampling_params' in kwargs:
+        return kwargs['sampling_params']
+    if len(args) > 1:
+        return args[1]
+    return None
+
+def common_chat_logic(scope, pricing_info, environment, application_name, metrics,
+    capture_message_content, disable_metrics, version, is_stream):
+    """
+    Process chat request and generate Telemetry
+    """
+
+    scope._end_time = time.time()
+    if len(scope._timestamps) > 1:
+        scope._tbt = calculate_tbt(scope._timestamps)
+
+    # Set base span attributes
+    scope._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+    scope._span.set_attribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SYSTEM, SemanticConvention.GEN_AI_SYSTEM_VLLM)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL, scope._request_model)
+    scope._span.set_attribute(SemanticConvention.SERVER_PORT, scope._server_port)
+    scope._span.set_attribute(SemanticConvention.SERVER_ADDRESS, scope._server_address)
+
+    # Handle inference configuration
+    inference_config = get_inference_config(scope._args, scope._kwargs)
+    if inference_config:
+        attributes = [
+            (SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY, 'frequency_penalty'),
+            (SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, 'max_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'),
+        ]
+
+        for attribute, key in attributes:
+            value = getattr(inference_config, key, None)
+            if value is not None:
+                scope._span.set_attribute(attribute, value)
+
+    scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL, scope._request_model)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE, "text")
+
+    # Set base span attributes (Extras)
+    scope._span.set_attribute(DEPLOYMENT_ENVIRONMENT, environment)
+    scope._span.set_attribute(SERVICE_NAME, application_name)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IS_STREAM, is_stream)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TBT, scope._tbt)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TTFT, scope._ttft)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION, version)
+
+    input_tokens = 0
+    output_tokens = 0
+    cost = 0
+
+    if capture_message_content:
+        prompt = ""
+        completion = ""
+
+        for output in scope._response:
+            prompt += output.prompt + "\n"
+            if output.outputs and len(output.outputs) > 0:
+                completion += output.outputs[0].text + "\n"
+            input_tokens += general_tokens(output.prompt)
+            output_tokens += general_tokens(output.outputs[0].text)
+
+        # Add a single event for prompt
+        scope._span.add_event(
+            name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
+            attributes={
+                SemanticConvention.GEN_AI_CONTENT_PROMPT: prompt,
+            },
+        )
+
+        # Add a single event for completion
+        scope._span.add_event(
+            name=SemanticConvention.GEN_AI_CONTENT_COMPLETION_EVENT,
+            attributes={
+                SemanticConvention.GEN_AI_CONTENT_COMPLETION: completion,
+            },
+        )
+
+    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
+                        input_tokens)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
+                        output_tokens)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE,
+                        input_tokens + output_tokens)
+
+    # Calculate cost of the operation
+    cost = get_chat_model_cost(scope._request_model, pricing_info, input_tokens, output_tokens)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
+
+    scope._span.set_status(Status(StatusCode.OK))
+
+    if disable_metrics is False:
+        metrics_attributes = create_metrics_attributes(
+            service_name=application_name,
+            deployment_environment=environment,
+            operation=SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+            system=SemanticConvention.GEN_AI_SYSTEM_VLLM,
+            request_model=scope._request_model,
+            server_address=scope._server_address,
+            server_port=scope._server_port,
+            response_model=scope._request_model,
+        )
+        metrics['genai_client_operation_duration'].record(scope._end_time - scope._start_time, metrics_attributes)
+        metrics['genai_server_tbt'].record(scope._tbt, metrics_attributes)
+        metrics['genai_server_ttft'].record(scope._ttft, metrics_attributes)
+        metrics['genai_requests'].add(1, metrics_attributes)
+        metrics['genai_completion_tokens'].add(output_tokens, metrics_attributes)
+        metrics['genai_prompt_tokens'].add(input_tokens, metrics_attributes)
+        metrics['genai_cost'].record(cost, metrics_attributes)
+        metrics['genai_client_usage_tokens'].record(
+            input_tokens + output_tokens, metrics_attributes)
+
+def process_chat_response(instance, response, request_model, pricing_info, server_port, server_address,
+    environment, application_name, metrics, start_time, span, args, kwargs,
+    capture_message_content=False, disable_metrics=False, version="1.0.0"):
+    """
+    Process chat request and generate Telemetry
+    """
+    self = type('GenericScope', (), {})()
+    self._response = response
+    self._start_time = start_time
+    self._end_time = time.time()
+    self._span = span
+    self._ttft, self._tbt = self._end_time - self._start_time, 0
+    self._server_address = server_address
+    self._server_port = server_port
+    self._request_model = request_model
+    self._timestamps = []
+    self._args = args
+    self._kwargs = kwargs
+
+    common_chat_logic(self, pricing_info, environment, application_name, metrics,
+        capture_message_content, disable_metrics, version, is_stream=False)
+
+    return response