openlit 1.34.18__tar.gz → 1.34.20__tar.gz

{openlit-1.34.18 → openlit-1.34.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: openlit
-Version: 1.34.18
+Version: 1.34.20
 Summary: OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects
 License: Apache-2.0
 Keywords: OpenTelemetry,otel,otlp,llm,tracing,openai,anthropic,claude,cohere,llm monitoring,observability,monitoring,gpt,Generative AI,chatGPT,gpu

{openlit-1.34.18 → openlit-1.34.20}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "openlit"
-version = "1.34.18"
+version = "1.34.20"
 description = "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects"
 authors = ["OpenLIT"]
 license = "Apache-2.0"

openlit-1.34.20/src/openlit/instrumentation/anthropic/__init__.py

@@ -0,0 +1,48 @@
+"""Initializer of Auto Instrumentation of Anthropic Functions"""
+
+from typing import Collection
+import importlib.metadata
+from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
+from wrapt import wrap_function_wrapper
+
+from openlit.instrumentation.anthropic.anthropic import messages
+from openlit.instrumentation.anthropic.async_anthropic import async_messages
+
+_instruments = ("anthropic >= 0.21.0",)
+
+class AnthropicInstrumentor(BaseInstrumentor):
+    """
+    An instrumentor for Anthropic's client library.
+    """
+
+    def instrumentation_dependencies(self) -> Collection[str]:
+        return _instruments
+
+    def _instrument(self, **kwargs):
+        version = importlib.metadata.version("anthropic")
+        environment = kwargs.get("environment", "default")
+        application_name = kwargs.get("application_name", "default")
+        tracer = kwargs.get("tracer")
+        pricing_info = kwargs.get("pricing_info", {})
+        capture_message_content = kwargs.get("capture_message_content", False)
+        metrics = kwargs.get("metrics_dict")
+        disable_metrics = kwargs.get("disable_metrics")
+
+        # sync
+        wrap_function_wrapper(
+            "anthropic.resources.messages",
+            "Messages.create",
+            messages(version, environment, application_name, tracer, pricing_info,
+                capture_message_content, metrics, disable_metrics),
+        )
+
+        # async
+        wrap_function_wrapper(
+            "anthropic.resources.messages",
+            "AsyncMessages.create",
+            async_messages(version, environment, application_name, tracer, pricing_info,
+                capture_message_content, metrics, disable_metrics),
+        )
+
+    def _uninstrument(self, **kwargs):
+        pass

openlit-1.34.20/src/openlit/instrumentation/anthropic/anthropic.py

@@ -0,0 +1,142 @@
+"""
+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.instrumentation.anthropic.utils import (
+    process_chunk,
+    process_chat_response,
+    process_streaming_chat_response,
+)
+from openlit.semcov import SemanticConvention
+
+def messages(version, environment, application_name, tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for Anthropic Messages.create calls.
+    """
+
+    class TracedSyncStream:
+        """
+        Wrapper for streaming responses to collect telemetry.
+        """
+
+        def __init__(
+                self,
+                wrapped,
+                span,
+                span_name,
+                kwargs,
+                server_address,
+                server_port,
+            ):
+            self.__wrapped__ = wrapped
+            self._span = span
+            self._span_name = span_name
+            self._llmresponse = ""
+            self._response_id = ""
+            self._response_model = ""
+            self._finish_reason = ""
+            self._input_tokens = 0
+            self._output_tokens = 0
+            self._tool_arguments = ""
+            self._tool_id = ""
+            self._tool_name = ""
+            self._tool_calls = None
+            self._response_role = ""
+            self._kwargs = kwargs
+            self._start_time = time.time()
+            self._end_time = None
+            self._timestamps = []
+            self._ttft = 0
+            self._tbt = 0
+            self._server_address = server_address
+            self._server_port = server_port
+
+        def __enter__(self):
+            self.__wrapped__.__enter__()
+            return self
+
+        def __exit__(self, exc_type, exc_value, traceback):
+            self.__wrapped__.__exit__(exc_type, exc_value, traceback)
+
+        def __iter__(self):
+            return self
+
+        def __getattr__(self, name):
+            """Delegate attribute access to the wrapped object."""
+            return getattr(self.__wrapped__, name)
+
+        def __next__(self):
+            try:
+                chunk = self.__wrapped__.__next__()
+                process_chunk(self, chunk)
+                return chunk
+            except StopIteration:
+                try:
+                    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
+                        )
+                except Exception as e:
+                    handle_exception(self._span, e)
+                raise
+
+    def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the Anthropic Messages.create call.
+        """
+
+        streaming = kwargs.get("stream", False)
+        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}"
+
+        # pylint: disable=no-else-return
+        if streaming:
+            awaited_wrapped = wrapped(*args, **kwargs)
+            span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
+
+            return TracedSyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port)
+
+        else:
+            with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+                start_time = time.time()
+                response = wrapped(*args, **kwargs)
+
+                try:
+                    response = process_chat_response(
+                        response=response,
+                        request_model=request_model,
+                        pricing_info=pricing_info,
+                        server_port=server_port,
+                        server_address=server_address,
+                        environment=environment,
+                        application_name=application_name,
+                        metrics=metrics,
+                        start_time=start_time,
+                        span=span,
+                        capture_message_content=capture_message_content,
+                        disable_metrics=disable_metrics,
+                        version=version,
+                        **kwargs
+                    )
+
+                except Exception as e:
+                    handle_exception(span, e)
+
+            return response
+
+    return wrapper

openlit-1.34.20/src/openlit/instrumentation/anthropic/async_anthropic.py

@@ -0,0 +1,142 @@
+"""
+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.instrumentation.anthropic.utils import (
+    process_chunk,
+    process_chat_response,
+    process_streaming_chat_response,
+)
+from openlit.semcov import SemanticConvention
+
+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.
+    """
+
+    class TracedAsyncStream:
+        """
+        Wrapper for async streaming responses to collect telemetry.
+        """
+
+        def __init__(
+                self,
+                wrapped,
+                span,
+                span_name,
+                kwargs,
+                server_address,
+                server_port,
+            ):
+            self.__wrapped__ = wrapped
+            self._span = span
+            self._span_name = span_name
+            self._llmresponse = ""
+            self._response_id = ""
+            self._response_model = ""
+            self._finish_reason = ""
+            self._input_tokens = 0
+            self._output_tokens = 0
+            self._tool_arguments = ""
+            self._tool_id = ""
+            self._tool_name = ""
+            self._tool_calls = None
+            self._response_role = ""
+            self._kwargs = kwargs
+            self._start_time = time.time()
+            self._end_time = None
+            self._timestamps = []
+            self._ttft = 0
+            self._tbt = 0
+            self._server_address = server_address
+            self._server_port = server_port
+
+        async def __aenter__(self):
+            await self.__wrapped__.__aenter__()
+            return self
+
+        async def __aexit__(self, exc_type, exc_value, traceback):
+            await self.__wrapped__.__aexit__(exc_type, exc_value, traceback)
+
+        def __aiter__(self):
+            return self
+
+        async def __getattr__(self, name):
+            """Delegate attribute access to the wrapped object."""
+            return getattr(await self.__wrapped__, name)
+
+        async def __anext__(self):
+            try:
+                chunk = await self.__wrapped__.__anext__()
+                process_chunk(self, chunk)
+                return chunk
+            except StopAsyncIteration:
+                try:
+                    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
+                        )
+                except Exception as e:
+                    handle_exception(self._span, e)
+                raise
+
+    async def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the Anthropic AsyncMessages.create call.
+        """
+
+        streaming = kwargs.get("stream", False)
+        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}"
+
+        # pylint: disable=no-else-return
+        if streaming:
+            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)
+
+        else:
+            with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+                start_time = time.time()
+                response = await wrapped(*args, **kwargs)
+
+                try:
+                    response = process_chat_response(
+                        response=response,
+                        request_model=request_model,
+                        pricing_info=pricing_info,
+                        server_port=server_port,
+                        server_address=server_address,
+                        environment=environment,
+                        application_name=application_name,
+                        metrics=metrics,
+                        start_time=start_time,
+                        span=span,
+                        capture_message_content=capture_message_content,
+                        disable_metrics=disable_metrics,
+                        version=version,
+                        **kwargs
+                    )
+
+                except Exception as e:
+                    handle_exception(span, e)
+
+            return response
+
+    return wrapper

openlit-1.34.20/src/openlit/instrumentation/anthropic/utils.py

@@ -0,0 +1,225 @@
+"""
+Anthropic OpenTelemetry instrumentation utility functions
+"""
+import time
+
+from opentelemetry.trace import Status, StatusCode
+
+from openlit.__helpers import (
+    calculate_ttft,
+    response_as_dict,
+    calculate_tbt,
+    get_chat_model_cost,
+    record_completion_metrics,
+    common_span_attributes,
+)
+from openlit.semcov import SemanticConvention
+
+def format_content(messages):
+    """
+    Format the messages into a string for span events.
+    """
+
+    if not messages:
+        return ""
+
+    formatted_messages = []
+    for message in messages:
+        if isinstance(message, dict):
+            role = message.get("role", "user")
+            content = message.get("content", "")
+        else:
+            # Handle Anthropic object format
+            role = getattr(message, "role", "user")
+            content = getattr(message, "content", "")
+
+        if isinstance(content, list):
+            # Handle structured content (e.g., text + images)
+            text_parts = []
+            for part in content:
+                if isinstance(part, dict) and part.get("type") == "text":
+                    text_parts.append(part.get("text", ""))
+            content = " ".join(text_parts)
+        elif not isinstance(content, str):
+            content = str(content)
+
+        formatted_messages.append(f"{role}: {content}")
+
+    return "\n".join(formatted_messages)
+
+def process_chunk(scope, chunk):
+    """
+    Process a chunk of response data and update state.
+    """
+
+    end_time = time.time()
+    # Record the timestamp for the current chunk
+    scope._timestamps.append(end_time)
+
+    if len(scope._timestamps) == 1:
+        # Calculate time to first chunk
+        scope._ttft = calculate_ttft(scope._timestamps, scope._start_time)
+
+    chunked = response_as_dict(chunk)
+
+    # Collect message IDs and input token from events
+    if chunked.get("type") == "message_start":
+        scope._response_id = chunked.get("message").get("id")
+        scope._input_tokens = chunked.get("message").get("usage").get("input_tokens")
+        scope._response_model = chunked.get("message").get("model")
+        scope._response_role = chunked.get("message").get("role")
+
+    # Collect message IDs and aggregated response from events
+    if chunked.get("type") == "content_block_delta":
+        if chunked.get("delta").get("text"):
+            scope._llmresponse += chunked.get("delta").get("text")
+        elif chunked.get("delta").get("partial_json"):
+            scope._tool_arguments += chunked.get("delta").get("partial_json")
+
+    if chunked.get("type") == "content_block_start":
+        if chunked.get("content_block").get("id"):
+            scope._tool_id = chunked.get("content_block").get("id")
+        if chunked.get("content_block").get("name"):
+            scope._tool_name = chunked.get("content_block").get("name")
+
+    # Collect output tokens and stop reason from events
+    if chunked.get("type") == "message_delta":
+        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):
+    """
+    Process chat request and generate Telemetry
+    """
+
+    scope._end_time = time.time()
+    if len(scope._timestamps) > 1:
+        scope._tbt = calculate_tbt(scope._timestamps)
+
+    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)
+
+    # 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)
+
+    # 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))
+
+    # 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")
+
+    # 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_COST, cost)
+
+    # Handle tool calls if present
+    if scope._tool_calls:
+        # Optimized tool handling - extract name, id, and arguments
+        tool_name = scope._tool_calls.get("name", "")
+        tool_id = scope._tool_calls.get("id", "")
+        tool_args = scope._tool_calls.get("input", "")
+
+        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_NAME, tool_name)
+        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_CALL_ID, tool_id)
+        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_ARGS, str(tool_args))
+
+    # 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)
+
+        # To be removed once the change to span_attributes (from span events) is complete
+        scope._span.add_event(
+            name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
+            attributes={
+                SemanticConvention.GEN_AI_CONTENT_PROMPT: formatted_messages,
+            },
+        )
+        scope._span.add_event(
+            name=SemanticConvention.GEN_AI_CONTENT_COMPLETION_EVENT,
+            attributes={
+                SemanticConvention.GEN_AI_CONTENT_COMPLETION: scope._llmresponse,
+            },
+        )
+
+    scope._span.set_status(Status(StatusCode.OK))
+
+    # 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)
+
+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.
+    """
+
+    if scope._tool_id != "":
+        scope._tool_calls = {
+            "id": scope._tool_id,
+            "name": scope._tool_name,
+            "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):
+    """
+    Process non-streaming chat response and generate telemetry.
+    """
+
+    scope = type("GenericScope", (), {})()
+    response_dict = response_as_dict(response)
+
+    # pylint: disable = no-member
+    scope._start_time = start_time
+    scope._end_time = time.time()
+    scope._span = span
+    scope._llmresponse = response_dict.get("content", [{}])[0].get("text", "")
+    scope._response_role = response_dict.get("role", "assistant")
+    scope._input_tokens = response_dict.get("usage").get("input_tokens")
+    scope._output_tokens = response_dict.get("usage").get("output_tokens")
+    scope._response_model = response_dict.get("model", "")
+    scope._finish_reason = response_dict.get("stop_reason", "")
+    scope._response_id = response_dict.get("id", "")
+    scope._timestamps = []
+    scope._ttft, scope._tbt = scope._end_time - scope._start_time, 0
+    scope._server_address, scope._server_port = server_address, server_port
+    scope._kwargs = kwargs
+
+    # Handle tool calls if present
+    content_blocks = response_dict.get("content", [])
+    scope._tool_calls = None
+    for block in content_blocks:
+        if block.get("type") == "tool_use":
+            scope._tool_calls = {
+                "id": block.get("id", ""),
+                "name": block.get("name", ""),
+                "input": block.get("input", "")
+            }
+            break
+
+    common_chat_logic(scope, pricing_info, environment, application_name, metrics,
+                        capture_message_content, disable_metrics, version, is_stream=False)
+
+    return response

{openlit-1.34.18 → openlit-1.34.20}/src/openlit/instrumentation/bedrock/__init__.py

@@ -1,4 +1,3 @@
-# pylint: disable=useless-return, bad-staticmethod-argument, disable=duplicate-code
 """Initializer of Auto Instrumentation of AWS Bedrock Functions"""
 
 from typing import Collection
@@ -6,37 +5,43 @@ import importlib.metadata
 from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
 from wrapt import wrap_function_wrapper
 
-from openlit.instrumentation.bedrock.bedrock import converse
+from openlit.instrumentation.bedrock.bedrock import converse, converse_stream
 
 _instruments = ("boto3 >= 1.34.138",)
 
 class BedrockInstrumentor(BaseInstrumentor):
     """
-    An instrumentor for AWS Bedrock's client library.
+    An instrumentor for AWS Bedrock client library.
     """
 
     def instrumentation_dependencies(self) -> Collection[str]:
         return _instruments
 
     def _instrument(self, **kwargs):
-        application_name = kwargs.get("application_name", "default_application")
-        environment = kwargs.get("environment", "default_environment")
+        version = importlib.metadata.version("boto3")
+        environment = kwargs.get("environment", "default")
+        application_name = kwargs.get("application_name", "default")
         tracer = kwargs.get("tracer")
-        event_provider = kwargs.get('event_provider')
-        metrics = kwargs.get("metrics_dict")
         pricing_info = kwargs.get("pricing_info", {})
         capture_message_content = kwargs.get("capture_message_content", False)
+        metrics = kwargs.get("metrics_dict")
         disable_metrics = kwargs.get("disable_metrics")
-        version = importlib.metadata.version("boto3")
 
-        #sync
+        # sync
+        wrap_function_wrapper(
+            "botocore.client",
+            "ClientCreator.create_client",
+            converse(version, environment, application_name, tracer, pricing_info,
+                capture_message_content, metrics, disable_metrics),
+        )
+
+        # streaming
         wrap_function_wrapper(
-            "botocore.client",  
-            "ClientCreator.create_client",  
-            converse(version, environment, application_name,
-                     tracer, event_provider, pricing_info, capture_message_content, metrics, disable_metrics),
+            "botocore.client",
+            "ClientCreator.create_client",
+            converse_stream(version, environment, application_name, tracer, pricing_info,
+                capture_message_content, metrics, disable_metrics),
         )
 
     def _uninstrument(self, **kwargs):
-        # Proper uninstrumentation logic to revert patched methods
         pass