openlit 1.33.9__py3-none-any.whl → 1.33.11__py3-none-any.whl

openlit/instrumentation/qdrant/__init__.py

@@ -273,7 +273,7 @@ class QdrantInstrumentor(BaseInstrumentor):
         tracer = kwargs.get("tracer")
         metrics = kwargs.get("metrics_dict")
         pricing_info = kwargs.get("pricing_info")
-        trace_content = kwargs.get("trace_content")
+        capture_message_content = kwargs.get("capture_message_content")
         disable_metrics = kwargs.get("disable_metrics")
         version = importlib.metadata.version("qdrant-client")
 
@@ -286,7 +286,7 @@ class QdrantInstrumentor(BaseInstrumentor):
                 wrap_package,
                 wrap_object,
                 wrapper(gen_ai_endpoint, version, environment, application_name,
-                 tracer, pricing_info, trace_content, metrics, disable_metrics),
+                 tracer, pricing_info, capture_message_content, metrics, disable_metrics),
             )
 
 

openlit/instrumentation/qdrant/async_qdrant.py

@@ -5,7 +5,7 @@ Module for monitoring Qdrant.
 
 import logging
 from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
 from openlit.__helpers import handle_exception
 from openlit.semcov import SemanticConvetion
 
@@ -25,7 +25,7 @@ def object_count(obj):
     return cnt
 
 def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
-                 tracer, pricing_info, trace_content, metrics, disable_metrics):
+                 tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """
     Creates a wrapper around a function call to trace and log its execution metrics.
 
@@ -39,7 +39,7 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
     - application_name (str): Name of the Langchain application.
     - tracer (opentelemetry.trace.Tracer): The tracer object used for OpenTelemetry tracing.
     - pricing_info (dict): Information about the pricing for internal metrics (currently not used).
-    - trace_content (bool): Flag indicating whether to trace the content of the response.
+    - capture_message_content (bool): Flag indicating whether to trace the content of the response.
 
     Returns:
     - function: A higher-order function that takes a function 'wrapped' and returns
@@ -73,9 +73,9 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
                 span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
                 span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
                                    gen_ai_endpoint)
-                span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+                span.set_attribute(DEPLOYMENT_ENVIRONMENT,
                                    environment)
-                span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+                span.set_attribute(SERVICE_NAME,
                                    application_name)
                 span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
                                    SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
@@ -241,11 +241,11 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
                     attributes = {
                         TELEMETRY_SDK_NAME:
                             "openlit",
-                        SemanticConvetion.GEN_AI_APPLICATION_NAME:
+                        SERVICE_NAME:
                             application_name,
                         SemanticConvetion.DB_SYSTEM:
                             SemanticConvetion.DB_SYSTEM_QDRANT,
-                        SemanticConvetion.GEN_AI_ENVIRONMENT:
+                        DEPLOYMENT_ENVIRONMENT:
                             environment,
                         SemanticConvetion.GEN_AI_OPERATION:
                             SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,

openlit/instrumentation/qdrant/qdrant.py

@@ -5,7 +5,7 @@ Module for monitoring Qdrant.
 
 import logging
 from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
 from openlit.__helpers import handle_exception
 from openlit.semcov import SemanticConvetion
 
@@ -25,7 +25,7 @@ def object_count(obj):
     return cnt
 
 def general_wrap(gen_ai_endpoint, version, environment, application_name,
-                 tracer, pricing_info, trace_content, metrics, disable_metrics):
+                 tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """
     Creates a wrapper around a function call to trace and log its execution metrics.
 
@@ -39,7 +39,7 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
     - application_name (str): Name of the Langchain application.
     - tracer (opentelemetry.trace.Tracer): The tracer object used for OpenTelemetry tracing.
     - pricing_info (dict): Information about the pricing for internal metrics (currently not used).
-    - trace_content (bool): Flag indicating whether to trace the content of the response.
+    - capture_message_content (bool): Flag indicating whether to trace the content of the response.
 
     Returns:
     - function: A higher-order function that takes a function 'wrapped' and returns
@@ -73,9 +73,9 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
                 span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
                 span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
                                    gen_ai_endpoint)
-                span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+                span.set_attribute(DEPLOYMENT_ENVIRONMENT,
                                    environment)
-                span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+                span.set_attribute(SERVICE_NAME,
                                    application_name)
                 span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
                                    SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
@@ -248,11 +248,11 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
                     attributes = {
                         TELEMETRY_SDK_NAME:
                             "openlit",
-                        SemanticConvetion.GEN_AI_APPLICATION_NAME:
+                        SERVICE_NAME:
                             application_name,
                         SemanticConvetion.DB_SYSTEM:
                             SemanticConvetion.DB_SYSTEM_QDRANT,
-                        SemanticConvetion.GEN_AI_ENVIRONMENT:
+                        DEPLOYMENT_ENVIRONMENT:
                             environment,
                         SemanticConvetion.GEN_AI_OPERATION:
                             SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,

openlit/instrumentation/reka/__init__.py

@@ -29,7 +29,7 @@ class RekaInstrumentor(BaseInstrumentor):
         tracer = kwargs.get("tracer")
         metrics = kwargs.get("metrics_dict")
         pricing_info = kwargs.get("pricing_info", {})
-        trace_content = kwargs.get("trace_content", False)
+        capture_message_content = kwargs.get("capture_message_content", False)
         disable_metrics = kwargs.get("disable_metrics")
         version = importlib.metadata.version("reka-api")
 
@@ -37,16 +37,16 @@ class RekaInstrumentor(BaseInstrumentor):
         wrap_function_wrapper(
             "reka.chat.client",
             "ChatClient.create",
-            chat("reka.chat", version, environment, application_name,
-                  tracer, pricing_info, trace_content, metrics, disable_metrics),
+            chat(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
         # async chat
         wrap_function_wrapper(
             "reka.chat.client",
             "AsyncChatClient.create",
-            async_chat("reka.chat", version, environment, application_name,
-                  tracer, pricing_info, trace_content, metrics, disable_metrics),
+            async_chat(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
     def _uninstrument(self, **kwargs):

openlit/instrumentation/reka/async_reka.py

@@ -1,33 +1,34 @@
-# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, possibly-used-before-assignment
 """
 Module for monitoring Reka API calls.
 """
 
 import logging
+import time
 from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
 from openlit.__helpers import (
-    handle_exception,
     get_chat_model_cost,
+    handle_exception,
+    create_metrics_attributes,
+    set_server_address_and_port
 )
 from openlit.semcov import SemanticConvetion
 
 # Initialize logger for logging potential issues and operations
 logger = logging.getLogger(__name__)
 
-def async_chat(gen_ai_endpoint, version, environment, application_name,
-                     tracer, pricing_info, trace_content, metrics, disable_metrics):
+def async_chat(version, environment, application_name,
+                     tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """
     Generates a telemetry wrapper for chat to collect metrics.
 
     Args:
-        gen_ai_endpoint: Endpoint identifier for logging and tracing.
         version: Version of the monitoring package.
         environment: Deployment environment (e.g., production, staging).
         application_name: Name of the application using the Reka API.
         tracer: OpenTelemetry tracer for creating spans.
         pricing_info: Information used for calculating the cost of Reka usage.
-        trace_content: Flag indicating whether to trace the actual content.
+        capture_message_content: Flag indicating whether to trace the actual content.
 
     Returns:
         A function that wraps the chat method to add telemetry.
@@ -50,8 +51,15 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
             The response from the original 'chat' method.
         """
 
-        with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+        server_address, server_port = set_server_address_and_port(instance, "api.reka.ai", 443)
+        request_model = kwargs.get("model", "reka-core-20240501")
+
+        span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+
+        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+            start_time = time.time()
             response = await wrapped(*args, **kwargs)
+            end_time = time.time()
 
             try:
                 # Format 'messages' into a single string
@@ -63,7 +71,6 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
 
                     if isinstance(content, list):
                         content_str = ", ".join(
-                            # pylint: disable=line-too-long
                             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
@@ -73,23 +80,69 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
                         formatted_messages.append(f"{role}: {content}")
                 prompt = "\n".join(formatted_messages)
 
-                # Set base span attribues
+                input_tokens = response.usage.input_tokens
+                output_tokens = response.usage.output_tokens
+
+                # Calculate cost of the operation
+                cost = get_chat_model_cost(request_model,
+                                            pricing_info, input_tokens, output_tokens)
+
+                # Set Span attributes (OTel Semconv)
                 span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
-                                    SemanticConvetion.GEN_AI_SYSTEM_REKAAI)
                 span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
                                     SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
-                span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
-                                    gen_ai_endpoint)
-                span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+                span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
+                                    SemanticConvetion.GEN_AI_SYSTEM_REKAAI)
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+                                    request_model)
+                span.set_attribute(SemanticConvetion.SERVER_PORT,
+                                    server_port)
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
+                                    kwargs.get("seed", ""))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
+                                    kwargs.get("max_tokens", -1))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_STOP_SEQUENCES,
+                                    kwargs.get("stop", []))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
+                                    kwargs.get("presence_penalty", 0.0))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
+                                    kwargs.get("temperature", 0.4))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_K,
+                                    kwargs.get("top_k", 1.0))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
+                                    kwargs.get("top_p", 1.0))
+                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
+                                    [response.responses[0].finish_reason])
+                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
+                                    response.id)
+                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+                                    response.model)
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+                                    input_tokens)
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
+                                    output_tokens)
+                span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+                                    server_address)
+                span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+                                    'text')
+
+                # Set Span attributes (Extra)
+                span.set_attribute(DEPLOYMENT_ENVIRONMENT,
                                     environment)
-                span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+                span.set_attribute(SERVICE_NAME,
                                     application_name)
-                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
-                                    kwargs.get("model", "reka-core"))
                 span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
                                     False)
-                if trace_content:
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
+                                    input_tokens + output_tokens)
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
+                                    cost)
+                span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TTFT,
+                                    end_time - start_time)
+                span.set_attribute(SemanticConvetion.GEN_AI_SDK_VERSION,
+                                    version)
+
+                if capture_message_content:
                     span.add_event(
                         name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
                         attributes={
@@ -99,51 +152,36 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
                     span.add_event(
                         name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
                         attributes={
-                            # pylint: disable=line-too-long
                             SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.responses[0].message.content,
                         },
                     )
 
-                prompt_tokens = response.usage.input_tokens
-                completion_tokens = response.usage.output_tokens
-                total_tokens = prompt_tokens + completion_tokens
-                # Calculate cost of the operation
-                cost = get_chat_model_cost(kwargs.get("model", "reka-core"),
-                                            pricing_info, prompt_tokens, completion_tokens)
-
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
-                                    prompt_tokens)
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
-                                    completion_tokens)
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
-                                    total_tokens)
-                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
-                                    [response.responses[0].finish_reason])
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
-                                    cost)
-
                 span.set_status(Status(StatusCode.OK))
 
                 if disable_metrics is False:
-                    attributes = {
-                        TELEMETRY_SDK_NAME:
-                            "openlit",
-                        SemanticConvetion.GEN_AI_APPLICATION_NAME:
-                            application_name,
-                        SemanticConvetion.GEN_AI_SYSTEM:
-                            SemanticConvetion.GEN_AI_SYSTEM_REKAAI,
-                        SemanticConvetion.GEN_AI_ENVIRONMENT:
-                            environment,
-                        SemanticConvetion.GEN_AI_OPERATION:
-                            SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
-                        SemanticConvetion.GEN_AI_REQUEST_MODEL:
-                            kwargs.get("model", "reka-core")
-                    }
+                    attributes = create_metrics_attributes(
+                        service_name=application_name,
+                        deployment_environment=environment,
+                        operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
+                        system=SemanticConvetion.GEN_AI_SYSTEM_REKAAI,
+                        request_model=request_model,
+                        server_address=server_address,
+                        server_port=server_port,
+                        response_model=response.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_total_tokens"].add(total_tokens, attributes)
-                    metrics["genai_completion_tokens"].add(completion_tokens, attributes)
-                    metrics["genai_prompt_tokens"].add(prompt_tokens, 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

openlit/instrumentation/reka/reka.py

@@ -1,33 +1,34 @@
-# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, possibly-used-before-assignment
 """
 Module for monitoring Reka API calls.
 """
 
 import logging
+import time
 from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
 from openlit.__helpers import (
-    handle_exception,
     get_chat_model_cost,
+    handle_exception,
+    create_metrics_attributes,
+    set_server_address_and_port
 )
 from openlit.semcov import SemanticConvetion
 
 # Initialize logger for logging potential issues and operations
 logger = logging.getLogger(__name__)
 
-def chat(gen_ai_endpoint, version, environment, application_name,
-                     tracer, pricing_info, trace_content, metrics, disable_metrics):
+def chat(version, environment, application_name,
+                     tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """
     Generates a telemetry wrapper for chat to collect metrics.
 
     Args:
-        gen_ai_endpoint: Endpoint identifier for logging and tracing.
         version: Version of the monitoring package.
         environment: Deployment environment (e.g., production, staging).
         application_name: Name of the application using the Reka API.
         tracer: OpenTelemetry tracer for creating spans.
         pricing_info: Information used for calculating the cost of Reka usage.
-        trace_content: Flag indicating whether to trace the actual content.
+        capture_message_content: Flag indicating whether to trace the actual content.
 
     Returns:
         A function that wraps the chat method to add telemetry.
@@ -50,8 +51,15 @@ def chat(gen_ai_endpoint, version, environment, application_name,
             The response from the original 'chat' method.
         """
 
-        with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+        server_address, server_port = set_server_address_and_port(instance, "api.reka.ai", 443)
+        request_model = kwargs.get("model", "reka-core-20240501")
+
+        span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+
+        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+            start_time = time.time()
             response = wrapped(*args, **kwargs)
+            end_time = time.time()
 
             try:
                 # Format 'messages' into a single string
@@ -63,7 +71,6 @@ def chat(gen_ai_endpoint, version, environment, application_name,
 
                     if isinstance(content, list):
                         content_str = ", ".join(
-                            # pylint: disable=line-too-long
                             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
@@ -73,23 +80,69 @@ def chat(gen_ai_endpoint, version, environment, application_name,
                         formatted_messages.append(f"{role}: {content}")
                 prompt = "\n".join(formatted_messages)
 
-                # Set base span attribues
+                input_tokens = response.usage.input_tokens
+                output_tokens = response.usage.output_tokens
+
+                # Calculate cost of the operation
+                cost = get_chat_model_cost(request_model,
+                                            pricing_info, input_tokens, output_tokens)
+
+                # Set Span attributes (OTel Semconv)
                 span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
-                span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
-                                    SemanticConvetion.GEN_AI_SYSTEM_REKAAI)
                 span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
                                     SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
-                span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
-                                    gen_ai_endpoint)
-                span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+                span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
+                                    SemanticConvetion.GEN_AI_SYSTEM_REKAAI)
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+                                    request_model)
+                span.set_attribute(SemanticConvetion.SERVER_PORT,
+                                    server_port)
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
+                                    kwargs.get("seed", ""))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
+                                    kwargs.get("max_tokens", -1))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_STOP_SEQUENCES,
+                                    kwargs.get("stop", []))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
+                                    kwargs.get("presence_penalty", 0.0))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
+                                    kwargs.get("temperature", 0.4))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_K,
+                                    kwargs.get("top_k", 1.0))
+                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
+                                    kwargs.get("top_p", 1.0))
+                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
+                                    [response.responses[0].finish_reason])
+                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
+                                    response.id)
+                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+                                    response.model)
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+                                    input_tokens)
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
+                                    output_tokens)
+                span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+                                    server_address)
+                span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+                                    'text')
+
+                # Set Span attributes (Extra)
+                span.set_attribute(DEPLOYMENT_ENVIRONMENT,
                                     environment)
-                span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+                span.set_attribute(SERVICE_NAME,
                                     application_name)
-                span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
-                                    kwargs.get("model", "reka-core"))
                 span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
                                     False)
-                if trace_content:
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
+                                    input_tokens + output_tokens)
+                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
+                                    cost)
+                span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TTFT,
+                                    end_time - start_time)
+                span.set_attribute(SemanticConvetion.GEN_AI_SDK_VERSION,
+                                    version)
+
+                if capture_message_content:
                     span.add_event(
                         name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
                         attributes={
@@ -99,51 +152,36 @@ def chat(gen_ai_endpoint, version, environment, application_name,
                     span.add_event(
                         name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
                         attributes={
-                            # pylint: disable=line-too-long
                             SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.responses[0].message.content,
                         },
                     )
 
-                prompt_tokens = response.usage.input_tokens
-                completion_tokens = response.usage.output_tokens
-                total_tokens = prompt_tokens + completion_tokens
-                # Calculate cost of the operation
-                cost = get_chat_model_cost(kwargs.get("model", "reka-core"),
-                                            pricing_info, prompt_tokens, completion_tokens)
-
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
-                                    prompt_tokens)
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
-                                    completion_tokens)
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
-                                    total_tokens)
-                span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
-                                    [response.responses[0].finish_reason])
-                span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
-                                    cost)
-
                 span.set_status(Status(StatusCode.OK))
 
                 if disable_metrics is False:
-                    attributes = {
-                        TELEMETRY_SDK_NAME:
-                            "openlit",
-                        SemanticConvetion.GEN_AI_APPLICATION_NAME:
-                            application_name,
-                        SemanticConvetion.GEN_AI_SYSTEM:
-                            SemanticConvetion.GEN_AI_SYSTEM_REKAAI,
-                        SemanticConvetion.GEN_AI_ENVIRONMENT:
-                            environment,
-                        SemanticConvetion.GEN_AI_OPERATION:
-                            SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
-                        SemanticConvetion.GEN_AI_REQUEST_MODEL:
-                            kwargs.get("model", "reka-core")
-                    }
+                    attributes = create_metrics_attributes(
+                        service_name=application_name,
+                        deployment_environment=environment,
+                        operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
+                        system=SemanticConvetion.GEN_AI_SYSTEM_REKAAI,
+                        request_model=request_model,
+                        server_address=server_address,
+                        server_port=server_port,
+                        response_model=response.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_total_tokens"].add(total_tokens, attributes)
-                    metrics["genai_completion_tokens"].add(completion_tokens, attributes)
-                    metrics["genai_prompt_tokens"].add(prompt_tokens, 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

openlit/instrumentation/together/__init__.py

@@ -29,7 +29,7 @@ class TogetherInstrumentor(BaseInstrumentor):
         tracer = kwargs.get("tracer")
         metrics = kwargs.get("metrics_dict")
         pricing_info = kwargs.get("pricing_info", {})
-        trace_content = kwargs.get("trace_content", False)
+        capture_message_content = kwargs.get("capture_message_content", False)
         disable_metrics = kwargs.get("disable_metrics")
         version = importlib.metadata.version("together")
 
@@ -37,32 +37,32 @@ class TogetherInstrumentor(BaseInstrumentor):
         wrap_function_wrapper(
             "together.resources.chat.completions",  
             "ChatCompletions.create",  
-            completion("together.chat.completions", version, environment, application_name,
-                  tracer, pricing_info, trace_content, metrics, disable_metrics),
+            completion(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
         # Image generate
         wrap_function_wrapper(
             "together.resources.images",  
             "Images.generate",  
-            image_generate("together.image.generate", version, environment, application_name,
-                  tracer, pricing_info, trace_content, metrics, disable_metrics),
+            image_generate(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
         # Chat completions
         wrap_function_wrapper(
             "together.resources.chat.completions",  
             "AsyncChatCompletions.create",  
-            async_completion("together.chat.completions", version, environment, application_name,
-                  tracer, pricing_info, trace_content, metrics, disable_metrics),
+            async_completion(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
         # Image generate
         wrap_function_wrapper(
             "together.resources.images",  
             "AsyncImages.generate",  
-            async_image_generate("together.image.generate", version, environment, application_name,
-                  tracer, pricing_info, trace_content, metrics, disable_metrics),
+            async_image_generate(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
     def _uninstrument(self, **kwargs):