openlit 1.34.30__py3-none-any.whl → 1.34.31__py3-none-any.whl

openlit/instrumentation/anthropic/async_anthropic.py

@@ -4,10 +4,7 @@ Module for monitoring Anthropic API calls.
 
 import time
 from opentelemetry.trace import SpanKind
-from openlit.__helpers import (
-    handle_exception,
-    set_server_address_and_port
-)
+from openlit.__helpers import handle_exception, set_server_address_and_port
 from openlit.instrumentation.anthropic.utils import (
     process_chunk,
     process_chat_response,
@@ -15,7 +12,17 @@ from openlit.instrumentation.anthropic.utils import (
 )
 from openlit.semcov import SemanticConvention
 
-def async_messages(version, environment, application_name, tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def async_messages(
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for Anthropic AsyncMessages.create calls.
     """
@@ -26,14 +33,14 @@ def async_messages(version, environment, application_name, tracer, pricing_info,
         """
 
         def __init__(
-                self,
-                wrapped,
-                span,
-                span_name,
-                kwargs,
-                server_address,
-                server_port,
-            ):
+            self,
+            wrapped,
+            span,
+            span_name,
+            kwargs,
+            server_address,
+            server_port,
+        ):
             self.__wrapped__ = wrapped
             self._span = span
             self._span_name = span_name
@@ -87,7 +94,7 @@ def async_messages(version, environment, application_name, tracer, pricing_info,
                             metrics=metrics,
                             capture_message_content=capture_message_content,
                             disable_metrics=disable_metrics,
-                            version=version
+                            version=version,
                         )
                 except Exception as e:
                     handle_exception(self._span, e)
@@ -99,7 +106,9 @@ def async_messages(version, environment, application_name, tracer, pricing_info,
         """
 
         streaming = kwargs.get("stream", False)
-        server_address, server_port = set_server_address_and_port(instance, "api.anthropic.com", 443)
+        server_address, server_port = set_server_address_and_port(
+            instance, "api.anthropic.com", 443
+        )
         request_model = kwargs.get("model", "claude-3-5-sonnet-latest")
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
@@ -109,7 +118,9 @@ def async_messages(version, environment, application_name, tracer, pricing_info,
             awaited_wrapped = await wrapped(*args, **kwargs)
             span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
 
-            return TracedAsyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port)
+            return TracedAsyncStream(
+                awaited_wrapped, span, span_name, kwargs, server_address, server_port
+            )
 
         else:
             with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
@@ -131,7 +142,7 @@ def async_messages(version, environment, application_name, tracer, pricing_info,
                         capture_message_content=capture_message_content,
                         disable_metrics=disable_metrics,
                         version=version,
-                        **kwargs
+                        **kwargs,
                     )
 
                 except Exception as e:

openlit/instrumentation/anthropic/utils.py

@@ -1,6 +1,7 @@
 """
 Anthropic OpenTelemetry instrumentation utility functions
 """
+
 import time
 
 from opentelemetry.trace import Status, StatusCode
@@ -15,6 +16,7 @@ from openlit.__helpers import (
 )
 from openlit.semcov import SemanticConvention
 
+
 def format_content(messages):
     """
     Format the messages into a string for span events.
@@ -47,6 +49,7 @@ def format_content(messages):
 
     return "\n".join(formatted_messages)
 
+
 def process_chunk(scope, chunk):
     """
     Process a chunk of response data and update state.
@@ -87,8 +90,18 @@ def process_chunk(scope, chunk):
         scope._output_tokens = chunked.get("usage").get("output_tokens")
         scope._finish_reason = chunked.get("delta").get("stop_reason")
 
-def common_chat_logic(scope, pricing_info, environment, application_name, metrics,
-                        capture_message_content, disable_metrics, version, is_stream):
+
+def common_chat_logic(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content,
+    disable_metrics,
+    version,
+    is_stream,
+):
     """
     Process chat request and generate Telemetry
     """
@@ -100,30 +113,68 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
     formatted_messages = format_content(scope._kwargs.get("messages", []))
     request_model = scope._kwargs.get("model", "claude-3-5-sonnet-latest")
 
-    cost = get_chat_model_cost(request_model, pricing_info, scope._input_tokens, scope._output_tokens)
+    cost = get_chat_model_cost(
+        request_model, pricing_info, scope._input_tokens, scope._output_tokens
+    )
 
     # Common Span Attributes
-    common_span_attributes(scope,
-        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_ANTHROPIC,
-        scope._server_address, scope._server_port, request_model, scope._response_model,
-        environment, application_name, is_stream, scope._tbt, scope._ttft, version)
+    common_span_attributes(
+        scope,
+        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+        SemanticConvention.GEN_AI_SYSTEM_ANTHROPIC,
+        scope._server_address,
+        scope._server_port,
+        request_model,
+        scope._response_model,
+        environment,
+        application_name,
+        is_stream,
+        scope._tbt,
+        scope._ttft,
+        version,
+    )
 
     # Span Attributes for Request parameters
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, scope._kwargs.get("max_tokens", -1))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES, scope._kwargs.get("stop_sequences", []))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, scope._kwargs.get("temperature", 1.0))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_K, scope._kwargs.get("top_k", 1.0))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P, scope._kwargs.get("top_p", 1.0))
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS,
+        scope._kwargs.get("max_tokens", -1),
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_STOP_SEQUENCES,
+        scope._kwargs.get("stop_sequences", []),
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_TEMPERATURE,
+        scope._kwargs.get("temperature", 1.0),
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_TOP_K, scope._kwargs.get("top_k", 1.0)
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_TOP_P, scope._kwargs.get("top_p", 1.0)
+    )
 
     # Span Attributes for Response parameters
     scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID, scope._response_id)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON, [scope._finish_reason])
-    scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE, "text" if isinstance(scope._llmresponse, str) else "json")
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON, [scope._finish_reason]
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_OUTPUT_TYPE,
+        "text" if isinstance(scope._llmresponse, str) else "json",
+    )
 
     # Span Attributes for Cost and Tokens
-    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, scope._input_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, scope._output_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE, scope._input_tokens + scope._output_tokens)
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, scope._input_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, scope._output_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE,
+        scope._input_tokens + scope._output_tokens,
+    )
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
 
     # Handle tool calls if present
@@ -139,8 +190,12 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
 
     # Span Attributes for Content
     if capture_message_content:
-        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, formatted_messages)
-        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._llmresponse)
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_CONTENT_PROMPT, formatted_messages
+        )
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._llmresponse
+        )
 
         # To be removed once the change to span_attributes (from span events) is complete
         scope._span.add_event(
@@ -160,13 +215,36 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
 
     # Record metrics
     if not disable_metrics:
-        record_completion_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_ANTHROPIC,
-            scope._server_address, scope._server_port, request_model, scope._response_model, environment,
-            application_name, scope._start_time, scope._end_time, scope._input_tokens, scope._output_tokens,
-            cost, scope._tbt, scope._ttft)
+        record_completion_metrics(
+            metrics,
+            SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+            SemanticConvention.GEN_AI_SYSTEM_ANTHROPIC,
+            scope._server_address,
+            scope._server_port,
+            request_model,
+            scope._response_model,
+            environment,
+            application_name,
+            scope._start_time,
+            scope._end_time,
+            scope._input_tokens,
+            scope._output_tokens,
+            cost,
+            scope._tbt,
+            scope._ttft,
+        )
 
-def process_streaming_chat_response(scope, pricing_info, environment, application_name, metrics,
-                                    capture_message_content=False, disable_metrics=False, version=""):
+
+def process_streaming_chat_response(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="",
+):
     """
     Process streaming chat response and generate telemetry.
     """
@@ -175,15 +253,38 @@ def process_streaming_chat_response(scope, pricing_info, environment, applicatio
         scope._tool_calls = {
             "id": scope._tool_id,
             "name": scope._tool_name,
-            "input": scope._tool_arguments
+            "input": scope._tool_arguments,
         }
 
-    common_chat_logic(scope, pricing_info, environment, application_name, metrics,
-                        capture_message_content, disable_metrics, version, is_stream=True)
-
-def process_chat_response(response, request_model, pricing_info, server_port, server_address,
-                          environment, application_name, metrics, start_time,
-                          span, capture_message_content=False, disable_metrics=False, version="1.0.0", **kwargs):
+    common_chat_logic(
+        scope,
+        pricing_info,
+        environment,
+        application_name,
+        metrics,
+        capture_message_content,
+        disable_metrics,
+        version,
+        is_stream=True,
+    )
+
+
+def process_chat_response(
+    response,
+    request_model,
+    pricing_info,
+    server_port,
+    server_address,
+    environment,
+    application_name,
+    metrics,
+    start_time,
+    span,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="1.0.0",
+    **kwargs,
+):
     """
     Process non-streaming chat response and generate telemetry.
     """
@@ -215,11 +316,20 @@ def process_chat_response(response, request_model, pricing_info, server_port, se
             scope._tool_calls = {
                 "id": block.get("id", ""),
                 "name": block.get("name", ""),
-                "input": block.get("input", "")
+                "input": block.get("input", ""),
             }
             break
 
-    common_chat_logic(scope, pricing_info, environment, application_name, metrics,
-                        capture_message_content, disable_metrics, version, is_stream=False)
+    common_chat_logic(
+        scope,
+        pricing_info,
+        environment,
+        application_name,
+        metrics,
+        capture_message_content,
+        disable_metrics,
+        version,
+        is_stream=False,
+    )
 
     return response

openlit/instrumentation/assemblyai/__init__.py

@@ -9,6 +9,7 @@ from openlit.instrumentation.assemblyai.assemblyai import transcribe
 
 _instruments = ("assemblyai >= 0.35.1",)
 
+
 class AssemblyAIInstrumentor(BaseInstrumentor):
     """
     An instrumentor for AssemblyAI client library.
@@ -31,8 +32,16 @@ class AssemblyAIInstrumentor(BaseInstrumentor):
         wrap_function_wrapper(
             "assemblyai.transcriber",
             "Transcriber.transcribe",
-            transcribe(version, environment, application_name,
-                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            transcribe(
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
     def _uninstrument(self, **kwargs):

openlit/instrumentation/assemblyai/assemblyai.py

@@ -12,8 +12,17 @@ from openlit.semcov import SemanticConvention
 # Initialize logger for logging potential issues and operations
 logger = logging.getLogger(__name__)
 
-def transcribe(version, environment, application_name,
-                 tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def transcribe(
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for AssemblyAI transcribe function call
     """
@@ -23,7 +32,9 @@ def transcribe(version, environment, application_name,
         Wraps the AssemblyAI transcribe function call.
         """
 
-        server_address, server_port = set_server_address_and_port(instance, "api.assemblyai.com", 443)
+        server_address, server_port = set_server_address_and_port(
+            instance, "api.assemblyai.com", 443
+        )
         request_model = kwargs.get("speech_model", "best")
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO} {request_model}"
@@ -48,7 +59,7 @@ def transcribe(version, environment, application_name,
                     capture_message_content=capture_message_content,
                     disable_metrics=disable_metrics,
                     version=version,
-                    **kwargs
+                    **kwargs,
                 )
 
             except Exception as e:

openlit/instrumentation/assemblyai/utils.py

@@ -1,9 +1,14 @@
 """
 AssemblyAI OpenTelemetry instrumentation utility functions
 """
+
 import time
 
-from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
+from opentelemetry.sdk.resources import (
+    SERVICE_NAME,
+    TELEMETRY_SDK_NAME,
+    DEPLOYMENT_ENVIRONMENT,
+)
 from opentelemetry.trace import Status, StatusCode
 
 from openlit.__helpers import (
@@ -12,14 +17,29 @@ from openlit.__helpers import (
 )
 from openlit.semcov import SemanticConvention
 
+
 def format_audio_url(audio_url):
     """
     Process audio URL input to extract content.
     """
     return str(audio_url) if audio_url else ""
 
-def common_span_attributes(scope, gen_ai_operation, gen_ai_system, server_address, server_port,
-    request_model, response_model, environment, application_name, is_stream, tbt, ttft, version):
+
+def common_span_attributes(
+    scope,
+    gen_ai_operation,
+    gen_ai_system,
+    server_address,
+    server_port,
+    request_model,
+    response_model,
+    environment,
+    application_name,
+    is_stream,
+    tbt,
+    ttft,
+    version,
+):
     """
     Set common span attributes for both chat and RAG operations.
     """
@@ -30,7 +50,9 @@ def common_span_attributes(scope, gen_ai_operation, gen_ai_system, server_addres
     scope._span.set_attribute(SemanticConvention.SERVER_ADDRESS, server_address)
     scope._span.set_attribute(SemanticConvention.SERVER_PORT, server_port)
     scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL, request_model)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL, scope._response_model)
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_RESPONSE_MODEL, scope._response_model
+    )
     scope._span.set_attribute(DEPLOYMENT_ENVIRONMENT, environment)
     scope._span.set_attribute(SERVICE_NAME, application_name)
     scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IS_STREAM, is_stream)
@@ -38,8 +60,21 @@ def common_span_attributes(scope, gen_ai_operation, gen_ai_system, server_addres
     scope._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TTFT, scope._ttft)
     scope._span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION, version)
 
-def record_audio_metrics(metrics, gen_ai_operation, gen_ai_system, server_address, server_port,
-    request_model, response_model, environment, application_name, start_time, end_time, cost):
+
+def record_audio_metrics(
+    metrics,
+    gen_ai_operation,
+    gen_ai_system,
+    server_address,
+    server_port,
+    request_model,
+    response_model,
+    environment,
+    application_name,
+    start_time,
+    end_time,
+    cost,
+):
     """
     Record audio metrics for the operation.
     """
@@ -58,8 +93,18 @@ def record_audio_metrics(metrics, gen_ai_operation, gen_ai_system, server_addres
     metrics["genai_requests"].add(1, attributes)
     metrics["genai_cost"].record(cost, attributes)
 
-def common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, application_name,
-    metrics, capture_message_content, disable_metrics, version):
+
+def common_audio_logic(
+    scope,
+    gen_ai_endpoint,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content,
+    disable_metrics,
+    version,
+):
     """
     Process audio transcription request and generate Telemetry
     """
@@ -69,13 +114,26 @@ def common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, applic
     is_stream = False
 
     # Calculate cost based on audio duration
-    cost = get_audio_model_cost(request_model, pricing_info, prompt, scope._response.audio_duration)
+    cost = get_audio_model_cost(
+        request_model, pricing_info, prompt, scope._response.audio_duration
+    )
 
     # Common Span Attributes
-    common_span_attributes(scope,
-        SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO, SemanticConvention.GEN_AI_SYSTEM_ASSEMBLYAI,
-        scope._server_address, scope._server_port, request_model, request_model,
-        environment, application_name, is_stream, scope._tbt, scope._ttft, version)
+    common_span_attributes(
+        scope,
+        SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO,
+        SemanticConvention.GEN_AI_SYSTEM_ASSEMBLYAI,
+        scope._server_address,
+        scope._server_port,
+        request_model,
+        request_model,
+        environment,
+        application_name,
+        is_stream,
+        scope._tbt,
+        scope._ttft,
+        version,
+    )
 
     # Span Attributes for Response parameters
     scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE, "text")
@@ -85,12 +143,16 @@ def common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, applic
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
 
     # Audio-specific span attributes
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_DURATION, scope._response.audio_duration)
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_AUDIO_DURATION, scope._response.audio_duration
+    )
 
     # Span Attributes for Content
     if capture_message_content:
         scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, prompt)
-        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._response.text)
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._response.text
+        )
 
         # To be removed once the change to span_attributes (from span events) is complete
         scope._span.add_event(
@@ -110,14 +172,38 @@ def common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, applic
 
     # Metrics
     if not disable_metrics:
-        record_audio_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO,
-            SemanticConvention.GEN_AI_SYSTEM_ASSEMBLYAI, scope._server_address, scope._server_port,
-            request_model, request_model, environment, application_name, scope._start_time,
-            scope._end_time, cost)
-
-def process_audio_response(response, gen_ai_endpoint, pricing_info, server_port, server_address,
-    environment, application_name, metrics, start_time, span, capture_message_content=False,
-    disable_metrics=False, version="1.0.0", **kwargs):
+        record_audio_metrics(
+            metrics,
+            SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO,
+            SemanticConvention.GEN_AI_SYSTEM_ASSEMBLYAI,
+            scope._server_address,
+            scope._server_port,
+            request_model,
+            request_model,
+            environment,
+            application_name,
+            scope._start_time,
+            scope._end_time,
+            cost,
+        )
+
+
+def process_audio_response(
+    response,
+    gen_ai_endpoint,
+    pricing_info,
+    server_port,
+    server_address,
+    environment,
+    application_name,
+    metrics,
+    start_time,
+    span,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="1.0.0",
+    **kwargs,
+):
     """
     Process audio transcription request and generate Telemetry
     """
@@ -136,7 +222,16 @@ def process_audio_response(response, gen_ai_endpoint, pricing_info, server_port,
     scope._tbt = 0.0
     scope._ttft = scope._end_time - scope._start_time
 
-    common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, application_name,
-        metrics, capture_message_content, disable_metrics, version)
+    common_audio_logic(
+        scope,
+        gen_ai_endpoint,
+        pricing_info,
+        environment,
+        application_name,
+        metrics,
+        capture_message_content,
+        disable_metrics,
+        version,
+    )
 
     return response