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

openlit/instrumentation/openai/openai.py

@@ -2,78 +2,55 @@
 Module for monitoring OpenAI 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,
-    get_embed_model_cost,
-    get_audio_model_cost,
-    get_image_model_cost,
-    general_tokens,
     handle_exception,
-    extract_and_format_input,
-    concatenate_all_contents,
-    response_as_dict,
-    calculate_ttft,
-    calculate_tbt,
-    create_metrics_attributes,
     set_server_address_and_port
 )
+from openlit.instrumentation.openai.utils import (
+    process_chat_chunk,
+    process_response_chunk,
+    process_chat_response,
+    process_streaming_chat_response,
+    process_streaming_response_response,
+    process_response_response,
+    process_embedding_response,
+    process_image_response,
+    process_audio_response,
+)
 from openlit.semcov import SemanticConvention
 
-# Initialize logger for logging potential issues and operations
-logger = logging.getLogger(__name__)
-
-def responses(version, environment, application_name,
-                     tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+def chat_completions(version, environment, application_name, tracer, pricing_info,
+    capture_message_content, metrics, disable_metrics):
     """
-    Generates a telemetry wrapper for chat completions 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 method to add telemetry.
+    Generates a telemetry wrapper for OpenAI chat completions.
     """
 
     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,
-                **args,
             ):
             self.__wrapped__ = wrapped
             self._span = span
-            # Placeholder for aggregating streaming response
+            self._span_name = span_name
             self._llmresponse = ""
             self._response_id = ""
             self._response_model = ""
             self._finish_reason = ""
-            self._input_tokens = ""
-            self._output_tokens = ""
-
-            self._args = args
+            self._system_fingerprint = ""
+            self._service_tier = "auto"
+            self._tools = None
             self._kwargs = kwargs
             self._start_time = time.time()
             self._end_time = None
@@ -100,383 +77,106 @@ def responses(version, environment, application_name,
         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)
-
-                chunked = response_as_dict(chunk)
-                # Collect message IDs and aggregated response from events
-                if chunked.get('type') == "response.output_text.delta":
-                    self._llmresponse += chunked.get('delta')
-                if chunked.get('type') == "response.completed":
-                    self._response_id = chunked.get('response').get('id')
-                    self._response_model = chunked.get('response').get('model')
-                    self._finish_reason = chunked.get('response').get('status')
-                    self._input_tokens = chunked.get('response').get('usage').get('input_tokens')
-                    self._output_tokens = chunked.get('response').get('usage').get('output_tokens')
+                process_chat_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)
-
-                    try:
-                        formatted_messages = extract_and_format_input(self._kwargs.get('input', ''))
-                        prompt = concatenate_all_contents(formatted_messages)
-                    except:
-                        prompt = self._kwargs.get('input', '')
-
-                    request_model = self._kwargs.get("model", "gpt-4o")
-
-                    # Calculate cost of the operation
-                    cost = get_chat_model_cost(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_OPENAI)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                        request_model)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_SEED,
-                                        str(self._kwargs.get("seed", "")))
-                    self._span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        self._server_port)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS,
-                                        str(self._kwargs.get("max_output_tokens", -1)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES,
-                                        str(self._kwargs.get("stop", [])))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE,
-                                        str(self._kwargs.get("temperature", 1.0)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P,
-                                        str(self._kwargs.get("top_p", 1.0)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON,
-                                        [self._finish_reason])
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID,
-                                        self._response_id)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                        self._response_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_USER,
-                                        self._kwargs.get("user", ""))
-                    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_OPENAI,
-                            request_model=request_model,
-                            server_address=self._server_address,
-                            server_port=self._server_port,
-                            response_model=self._response_model,
+                    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_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
-                        )
-                        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 'chat.completions' 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' method.
+        Wraps the OpenAI chat completions call.
         """
 
-        # Check if streaming is enabled for the API call
         streaming = kwargs.get("stream", False)
         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}"
 
-        # 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)
+            return TracedSyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port)
 
-        # Handling for non-streaming responses
         else:
-            with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+            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()
-
-                response_dict = response_as_dict(response)
 
                 try:
-                    try:
-                        formatted_messages = extract_and_format_input(kwargs.get('input', ''))
-                        prompt = concatenate_all_contents(formatted_messages)
-                    except:
-                        prompt = kwargs.get('input', '')
-
-                    input_tokens = response_dict.get('usage').get('input_tokens')
-                    output_tokens = response_dict.get('usage').get('output_tokens')
-
-                    # 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_OPENAI)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                        request_model)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_SEED,
-                                        kwargs.get("seed", ""))
-                    span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        server_port)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS,
-                                        kwargs.get("max_output_tokens", -1))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES,
-                                        kwargs.get("stop", []))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE,
-                                        str(response_dict.get("temperature", 1.0)))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P,
-                                        str(response_dict.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)
-
-                    # 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_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,
-                            },
-                        )
-
-                    for i in range(kwargs.get('n',1)):
-                        span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON,
-                                           [response_dict.get('status')])
-                        try:
-                            llm_response = response_dict.get('output')[i].get('content')[0].get('text','')
-                        except:
-                            llm_response = ''
-
-                        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: llm_response,
-                                },
-                            )
-                        if kwargs.get('tools'):
-                            span.set_attribute(SemanticConvention.GEN_AI_TOOL_CALLS,
-                                            str(response_dict.get('tools')))
-
-                        if isinstance(llm_response, str):
-                            span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                            "text")
-                        elif llm_response 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_OPENAI,
+                    response = process_chat_response(
+                        response=response,
                             request_model=request_model,
-                            server_address=server_address,
+                        pricing_info=pricing_info,
                             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 original response
-                    return response
+                        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
 
-def chat_completions(version, environment, application_name,
-                     tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+def responses(version, environment, application_name, tracer, pricing_info,
+    capture_message_content, metrics, disable_metrics, **kwargs):
     """
-    Generates a telemetry wrapper for chat completions 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 method to add telemetry.
+    Generates a telemetry wrapper for OpenAI responses API.
     """
 
     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,
-                **args,
             ):
             self.__wrapped__ = wrapped
             self._span = span
-            # Placeholder for aggregating streaming response
+            self._span_name = span_name
             self._llmresponse = ""
             self._response_id = ""
             self._response_model = ""
             self._finish_reason = ""
-            self._openai_response_service_tier = ""
-            self._openai_system_fingerprint = ""
-
-            self._args = args
+            self._input_tokens = 0
+            self._output_tokens = 0
+            self._reasoning_tokens = 0
+            self._operation_type = "responses"
+            self._service_tier = "default"
+            self._tools = None
+            self._response_tools = None
             self._kwargs = kwargs
             self._start_time = time.time()
             self._end_time = None
@@ -503,578 +203,126 @@ def chat_completions(version, environment, application_name,
         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)
-
-                chunked = response_as_dict(chunk)
-                # Collect message IDs and aggregated response from events
-                if (len(chunked.get('choices')) > 0 and ('delta' in chunked.get('choices')[0] and
-                    'content' in chunked.get('choices')[0].get('delta'))):
-
-                    content = chunked.get('choices')[0].get('delta').get('content')
-                    if content:
-                        self._llmresponse += content
-                self._response_id = chunked.get('id')
-                self._response_model = chunked.get('model')
-                try:
-                    self._finish_reason = chunked.get('choices', [])[0].get('finish_reason')
-                except (IndexError, AttributeError, TypeError):
-                    self._finish_reason = "stop"
-                self._openai_response_service_tier = chunked.get('service_tier') or 'auto'
-                self._openai_system_fingerprint = chunked.get('system_fingerprint')
+                process_response_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)
-
-                    request_model = self._kwargs.get("model", "gpt-4o")
-
-                    # Calculate tokens using input prompt and aggregated response
-                    input_tokens = general_tokens(prompt)
-                    output_tokens = general_tokens(self._llmresponse)
-
-                    # Calculate cost of the operation
-                    cost = get_chat_model_cost(request_model,
-                                                pricing_info, input_tokens,
-                                                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_OPENAI)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                        request_model)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_SEED,
-                                        str(self._kwargs.get("seed", "")))
-                    self._span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        self._server_port)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY,
-                                        str(self._kwargs.get("frequency_penalty", 0.0)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS,
-                                        self._kwargs.get("max_tokens", -1))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY,
-                                        str(self._kwargs.get("presence_penalty", 0.0)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES,
-                                        str(self._kwargs.get("stop", [])))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE,
-                                        str(self._kwargs.get("temperature", 1.0)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P,
-                                        str(self._kwargs.get("top_p", 1.0)))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON,
-                                        [self._finish_reason])
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID,
-                                        self._response_id)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                        self._response_model)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
-                                        input_tokens)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
-                                        output_tokens)
-                    self._span.set_attribute(SemanticConvention.SERVER_ADDRESS,
-                                        self._server_address)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_SERVICE_TIER,
-                                        str(self._kwargs.get("service_tier", "auto")))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_SERVICE_TIER,
-                                        self._openai_response_service_tier)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_SYSTEM_FINGERPRINT,
-                                        self._openai_system_fingerprint)
-                    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_USER,
-                                        str(self._kwargs.get("user", "")))
-                    self._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IS_STREAM,
-                                        True)
-                    self._span.set_attribute(SemanticConvention.GEN_AI_USAGE_TOTAL_TOKENS,
-                                        input_tokens + 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,
-                            },
+                    with self._span:
+                        process_streaming_response_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
                         )
-                        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_OPENAI,
-                            request_model=request_model,
-                            server_address=self._server_address,
-                            server_port=self._server_port,
-                            response_model=self._response_model,
-                        )
-
-                        metrics["genai_client_usage_tokens"].record(
-                            input_tokens + 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
-                        )
-                        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)
-
                 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 'chat.completions' 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' method.
+        Wraps the OpenAI responses API call.
         """
 
-        # Check if streaming is enabled for the API call
         streaming = kwargs.get("stream", False)
         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}"
 
-        # 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)
+            return TracedSyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port)
 
-        # Handling for non-streaming responses
         else:
-            with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+            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()
-
-                response_dict = response_as_dict(response)
 
                 try:
-                    # Format 'messages' into a single string
-                    message_prompt = kwargs.get("messages", "")
-                    formatted_messages = []
-                    for message in message_prompt:
-                        role = message["role"]
-                        content = message["content"]
-
-                        if isinstance(content, list):
-                            content_str = ", ".join(
-                                f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
-                                if "type" in item else f'text: {item["text"]}'
-                                for item in content
-                            )
-                            formatted_messages.append(f"{role}: {content_str}")
-                        else:
-                            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 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_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_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,
-                            },
-                        )
-
-                    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 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_OPENAI,
+                    response = process_response_response(
+                        response=response,
                             request_model=request_model,
-                            server_address=server_address,
+                        pricing_info=pricing_info,
                             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 original response
-                    return response
+                        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
 
-def chat_completions_parse(version, environment, application_name, tracer, pricing_info, capture_message_content,
-                           metrics, disable_metrics):
+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.
+    Generates a telemetry wrapper for OpenAI chat completions parse.
     """
 
     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.
+        Wraps the OpenAI chat completions parse call.
         """
+
         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},
-                    )
+            response = wrapped(*args, **kwargs)
 
-                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,
+            try:
+                response = process_chat_response(
+                    response=response,
                         request_model=request_model,
-                        server_address=server_address,
+                    pricing_info=pricing_info,
                         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
+                    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 'parse' trace creation: %s", e)
-                # Re-raise the exception to not interfere with the application flow
-                raise
+
+            return response
 
     return wrapper
 
-def embedding(version, environment, application_name,
-              tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+def embedding(version, environment, application_name, tracer, pricing_info,
+    capture_message_content, metrics, disable_metrics, **kwargs):
     """
-    Generates a telemetry wrapper for embeddings 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 embeddings method to add telemetry.
+    Generates a telemetry wrapper for OpenAI embeddings.
     """
 
     def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the 'embeddings' 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 'embeddings' method to be wrapped.
-            instance: The instance of the class where the original method is defined.
-            args: Positional arguments for the 'embeddings' method.
-            kwargs: Keyword arguments for the 'embeddings' method.
-
-        Returns:
-            The response from the original 'embeddings' method.
+        Wraps the OpenAI embeddings call.
         """
 
         server_address, server_port = set_server_address_and_port(instance, "api.openai.com", 443)
@@ -1082,127 +330,44 @@ def embedding(version, environment, application_name,
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}"
 
-        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+        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()
 
-            response_dict = response_as_dict(response)
             try:
-                input_tokens = response_dict.get('usage').get('prompt_tokens')
-
-                # Calculate cost of the operation
-                cost = get_embed_model_cost(request_model,
-                                    pricing_info, input_tokens)
-
-                # Set Span attributes (OTel Semconv)
-                span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                    SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING)
-                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_ENCODING_FORMATS,
-                                    [kwargs.get('encoding_format', 'float')])
-                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                    request_model)
-                span.set_attribute(SemanticConvention.SERVER_ADDRESS,
-                                    server_address)
-                span.set_attribute(SemanticConvention.SERVER_PORT,
-                                    server_port)
-                span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
-                                    input_tokens)
-
-                # Set Span attributes (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_USAGE_TOTAL_TOKENS,
-                                    input_tokens)
-                span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                    cost)
-                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: str(kwargs.get("input", "")),
-                        },
-                    )
-
-                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_EMBEDDING,
-                        system=SemanticConvention.GEN_AI_SYSTEM_OPENAI,
+                response = process_embedding_response(
+                    response=response,
                         request_model=request_model,
-                        server_address=server_address,
+                    pricing_info=pricing_info,
                         server_port=server_port,
-                        response_model=request_model,
-                    )
-                    metrics["genai_client_usage_tokens"].record(
-                            input_tokens, attributes
-                        )
-                    metrics["genai_client_operation_duration"].record(
-                        end_time - start_time, attributes
-                    )
-                    metrics["genai_requests"].add(1, attributes)
-                    metrics["genai_prompt_tokens"].add(input_tokens, attributes)
-                    metrics["genai_cost"].record(cost, attributes)
-
-                # Return original response
-                return response
+                    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
 
-def image_generate(version, environment, application_name,
-                   tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+def image_generate(version, environment, application_name, tracer, pricing_info,
+    capture_message_content, metrics, disable_metrics, **kwargs):
     """
-    Generates a telemetry wrapper for image generation 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 image generation.
-        capture_message_content: Flag indicating whether to trace the input prompt and generated images.
-    
-    Returns:
-        A function that wraps the image generation method to add telemetry.
+    Generates a telemetry wrapper for OpenAI image generation.
     """
 
     def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the 'images.generate' API call to add telemetry.
-
-        This collects metrics such as execution time, cost, and handles errors
-        gracefully, adding details to the trace for observability.
-
-        Args:
-            wrapped: The original 'images.generate' method to be wrapped.
-            instance: The instance of the class where the original method is defined.
-            args: Positional arguments for the 'images.generate' method.
-            kwargs: Keyword arguments for the 'images.generate' method.
-
-        Returns:
-            The response from the original 'images.generate' method.
+        Wraps the OpenAI image generation call.
         """
 
         server_address, server_port = set_server_address_and_port(instance, "api.openai.com", 443)
@@ -1210,146 +375,46 @@ def image_generate(version, environment, application_name,
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_IMAGE} {request_model}"
 
-        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+        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()
 
-            images_count = 0
-
             try:
-                # Find Image format
-                if "response_format" in kwargs and kwargs["response_format"] == "b64_json":
-                    image = "b64_json"
-                else:
-                    image = "url"
-
-                # Calculate cost of the operation
-                cost = get_image_model_cost(request_model,
-                                            pricing_info, kwargs.get("size", "1024x1024"),
-                                            kwargs.get("quality", "standard"))
-
-                for items in response.data:
-                    # Set Span attributes (OTel Semconv)
-                    span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                    span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                        SemanticConvention.GEN_AI_OPERATION_TYPE_IMAGE)
-                    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.SERVER_ADDRESS,
-                                        server_address)
-                    span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        server_port)
-                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID,
-                                        response.created)
-                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                        request_model)
-                    span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                        "image")
-
-                    # Set Span attributes (Extras)
-                    span.set_attribute(DEPLOYMENT_ENVIRONMENT,
-                                        environment)
-                    span.set_attribute(SERVICE_NAME,
-                                        application_name)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IMAGE_SIZE,
-                                        kwargs.get("size", "1024x1024"))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IMAGE_QUALITY,
-                                        kwargs.get("quality", "standard"))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IMAGE_STYLE,
-                                        kwargs.get("style", "vivid"))
-                    span.set_attribute(SemanticConvention.GEN_AI_CONTENT_REVISED_PROMPT,
-                                        items.revised_prompt if items.revised_prompt else "")
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_USER,
-                                        kwargs.get("user", ""))
-                    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: kwargs.get("prompt", ""),
-                            },
-                        )
-                        attribute_name = f"{SemanticConvention.GEN_AI_RESPONSE_IMAGE}.{images_count}"
-                        span.add_event(
-                            name=attribute_name,
-                            attributes={
-                                SemanticConvention.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
-                            },
-                        )
-
-                    images_count+=1
-
-                span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                    len(response.data) * cost)
-                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_IMAGE,
-                        system=SemanticConvention.GEN_AI_SYSTEM_OPENAI,
+                response = process_image_response(
+                    response=response,
                         request_model=request_model,
-                        server_address=server_address,
+                    pricing_info=pricing_info,
                         server_port=server_port,
-                        response_model=request_model,
-                    )
-
-                    metrics["genai_client_operation_duration"].record(
-                        end_time - start_time, attributes
-                    )
-                    metrics["genai_requests"].add(1, attributes)
-                    metrics["genai_cost"].record(cost, attributes)
-
-                # Return original response
-                return response
+                    server_address=server_address,
+                    environment=environment,
+                    application_name=application_name,
+                    metrics=metrics,
+                    start_time=start_time,
+                    end_time=end_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
 
-def image_variatons(version, environment, application_name,
-                    tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+def image_variatons(version, environment, application_name, tracer, pricing_info,
+    capture_message_content, metrics, disable_metrics):
     """
-    Generates a telemetry wrapper for creating image variations 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 generating image variations.
-        capture_message_content: Flag indicating whether to trace the input image and generated variations.
-    
-    Returns:
-        A function that wraps the image variations creation method to add telemetry.
+    Generates a telemetry wrapper for OpenAI image variations.
     """
 
     def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the 'images.create.variations' API call to add telemetry.
-
-        This collects metrics such as execution time, cost, and handles errors
-        gracefully, adding details to the trace for observability.
-
-        Args:
-            wrapped: The original 'images.create.variations' method to be wrapped.
-            instance: The instance of the class where the original method is defined.
-            args: Positional arguments for the method.
-            kwargs: Keyword arguments for the method.
-
-        Returns:
-            The response from the original 'images.create.variations' method.
+        Wraps the OpenAI image variations call.
         """
 
         server_address, server_port = set_server_address_and_port(instance, "api.openai.com", 443)
@@ -1357,135 +422,46 @@ def image_variatons(version, environment, application_name,
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_IMAGE} {request_model}"
 
-        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+        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()
 
-            images_count = 0
-
             try:
-                # Find Image format
-                if "response_format" in kwargs and kwargs["response_format"] == "b64_json":
-                    image = "b64_json"
-                else:
-                    image = "url"
-
-                # Calculate cost of the operation
-                cost = get_image_model_cost(request_model, pricing_info,
-                                            kwargs.get("size", "1024x1024"), "standard")
-
-                for items in response.data:
-                    # Set Span attributes (OTel Semconv)
-                    span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                    span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                        SemanticConvention.GEN_AI_OPERATION_TYPE_IMAGE)
-                    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.SERVER_ADDRESS,
-                                        server_address)
-                    span.set_attribute(SemanticConvention.SERVER_PORT,
-                                        server_port)
-                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID,
-                                        response.created)
-                    span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                        request_model)
-                    span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                        "image")
-
-                    # Set Span attributes (Extras)
-                    span.set_attribute(DEPLOYMENT_ENVIRONMENT,
-                                        environment)
-                    span.set_attribute(SERVICE_NAME,
-                                        application_name)
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IMAGE_SIZE,
-                                        kwargs.get("size", "1024x1024"))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IMAGE_QUALITY,
-                                        "standard")
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_USER,
-                                        kwargs.get("user", ""))
-                    span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION,
-                                    version)
-
-                    if capture_message_content:
-                        attribute_name = f"{SemanticConvention.GEN_AI_RESPONSE_IMAGE}.{images_count}"
-                        span.add_event(
-                            name=attribute_name,
-                            attributes={
-                                SemanticConvention.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
-                            },
-                        )
-
-                    images_count+=1
-
-                span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                    len(response.data) * cost)
-                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_IMAGE,
-                        system=SemanticConvention.GEN_AI_SYSTEM_OPENAI,
+                response = process_image_response(
+                    response=response,
                         request_model=request_model,
-                        server_address=server_address,
+                    pricing_info=pricing_info,
                         server_port=server_port,
-                        response_model=request_model,
-                    )
-
-                    metrics["genai_client_operation_duration"].record(
-                        end_time - start_time, attributes
-                    )
-                    metrics["genai_requests"].add(1, attributes)
-                    metrics["genai_cost"].record(cost, attributes)
-
-                # Return original response
-                return response
+                    server_address=server_address,
+                    environment=environment,
+                    application_name=application_name,
+                    metrics=metrics,
+                    start_time=start_time,
+                    end_time=end_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
 
-def audio_create(version, environment, application_name,
-                 tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+def audio_create(version, environment, application_name, tracer, pricing_info,
+    capture_message_content, metrics, disable_metrics):
     """
-    Generates a telemetry wrapper for creating speech audio 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 generating speech audio.
-        capture_message_content: Flag indicating whether to trace the input text and generated audio.
-    
-    Returns:
-        A function that wraps the speech audio creation method to add telemetry.
+    Generates a telemetry wrapper for OpenAI audio creation.
     """
 
     def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the 'audio.speech.create' API call to add telemetry.
-
-        This collects metrics such as execution time, cost, and handles errors
-        gracefully, adding details to the trace for observability.
-
-        Args:
-            wrapped: The original 'audio.speech.create' method to be wrapped.
-            instance: The instance of the class where the original method is defined.
-            args: Positional arguments for the 'audio.speech.create' method.
-            kwargs: Keyword arguments for the 'audio.speech.create' method.
-
-        Returns:
-            The response from the original 'audio.speech.create' method.
+        Wraps the OpenAI audio creation call.
         """
 
         server_address, server_port = set_server_address_and_port(instance, "api.openai.com", 443)
@@ -1499,77 +475,27 @@ def audio_create(version, environment, application_name,
             end_time = time.time()
 
             try:
-                # Calculate cost of the operation
-                cost = get_audio_model_cost(request_model,
-                                            pricing_info, kwargs.get("input", ""))
-
-                # Set Span attributes
-                span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                    SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO)
-                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.SERVER_ADDRESS,
-                                    server_address)
-                span.set_attribute(SemanticConvention.SERVER_PORT,
-                                    server_port)
-                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                    request_model)
-                span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                    "speech")
-
-                span.set_attribute(DEPLOYMENT_ENVIRONMENT,
-                                    environment)
-                span.set_attribute(SERVICE_NAME,
-                                    application_name)
-                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_VOICE,
-                                    kwargs.get("voice", "alloy"))
-                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT,
-                                    kwargs.get("response_format", "mp3"))
-                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_SPEED,
-                                    kwargs.get("speed", 1))
-                span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                    cost)
-                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: kwargs.get("input", ""),
-                        },
-                    )
-
-                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_AUDIO,
-                        system=SemanticConvention.GEN_AI_SYSTEM_OPENAI,
+                response = process_audio_response(
+                    response=response,
                         request_model=request_model,
-                        server_address=server_address,
+                    pricing_info=pricing_info,
                         server_port=server_port,
-                        response_model=request_model,
-                    )
-
-                    metrics["genai_client_operation_duration"].record(
-                        end_time - start_time, attributes
-                    )
-                    metrics["genai_requests"].add(1, attributes)
-                    metrics["genai_cost"].record(cost, attributes)
-
-                # Return original response
-                return response
+                    server_address=server_address,
+                    environment=environment,
+                    application_name=application_name,
+                    metrics=metrics,
+                    start_time=start_time,
+                    end_time=end_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