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

openlit/instrumentation/gpt4all/gpt4all.py

@@ -4,20 +4,26 @@ Module for monitoring GPT4All 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.gpt4all.utils import (
     process_generate_response,
     process_chunk,
     process_streaming_generate_response,
-    process_embedding_response
+    process_embedding_response,
 )
 from openlit.semcov import SemanticConvention
 
-def generate(version, environment, application_name,
-    tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def generate(
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for GenAI function call
     """
@@ -28,16 +34,16 @@ def generate(version, environment, application_name,
         """
 
         def __init__(
-                self,
-                wrapped,
-                span,
-                span_name,
-                args,
-                kwargs,
-                server_address,
-                server_port,
-                request_model,
-            ):
+            self,
+            wrapped,
+            span,
+            span_name,
+            args,
+            kwargs,
+            server_address,
+            server_port,
+            request_model,
+        ):
             self.__wrapped__ = wrapped
             self._span = span
             self._span_name = span_name
@@ -75,7 +81,9 @@ def generate(version, environment, application_name,
                 return chunk
             except StopIteration:
                 try:
-                    with tracer.start_as_current_span(self._span_name, kind=SpanKind.CLIENT) as self._span:
+                    with tracer.start_as_current_span(
+                        self._span_name, kind=SpanKind.CLIENT
+                    ) as self._span:
                         process_streaming_generate_response(
                             self,
                             pricing_info=pricing_info,
@@ -84,7 +92,7 @@ def generate(version, environment, application_name,
                             metrics=metrics,
                             capture_message_content=capture_message_content,
                             disable_metrics=disable_metrics,
-                            version=version
+                            version=version,
                         )
 
                 except Exception as e:
@@ -100,8 +108,13 @@ def generate(version, environment, application_name,
         # Check if streaming is enabled for the API call
         streaming = kwargs.get("streaming", False)
 
-        server_address, server_port = set_server_address_and_port(instance, "127.0.0.1", 80)
-        request_model = str(instance.model.model_path).rsplit("/", maxsplit=1)[-1] or "orca-mini-3b-gguf2-q4_0.gguf"
+        server_address, server_port = set_server_address_and_port(
+            instance, "127.0.0.1", 80
+        )
+        request_model = (
+            str(instance.model.model_path).rsplit("/", maxsplit=1)[-1]
+            or "orca-mini-3b-gguf2-q4_0.gguf"
+        )
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
 
@@ -109,7 +122,16 @@ def generate(version, environment, application_name,
             # 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, span_name, args, kwargs, server_address, server_port, request_model)
+            return TracedSyncStream(
+                awaited_wrapped,
+                span,
+                span_name,
+                args,
+                kwargs,
+                server_address,
+                server_port,
+                request_model,
+            )
 
         # Handling for non-streaming responses
         else:
@@ -133,7 +155,7 @@ def generate(version, environment, application_name,
                         kwargs=kwargs,
                         capture_message_content=capture_message_content,
                         disable_metrics=disable_metrics,
-                        version=version
+                        version=version,
                     )
 
                 except Exception as e:
@@ -143,8 +165,17 @@ def generate(version, environment, application_name,
 
     return wrapper
 
-def embed(version, environment, application_name,
-    tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def embed(
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for GenAI function call
     """
@@ -154,10 +185,17 @@ def embed(version, environment, application_name,
         Wraps the GenAI function call.
         """
 
-        server_address, server_port = set_server_address_and_port(instance, "127.0.0.1", 80)
-        request_model = str(instance.gpt4all.model.model_path).rsplit("/", maxsplit=1)[-1] or "all-MiniLM-L6-v2.gguf2.f16.gguf"
+        server_address, server_port = set_server_address_and_port(
+            instance, "127.0.0.1", 80
+        )
+        request_model = (
+            str(instance.gpt4all.model.model_path).rsplit("/", maxsplit=1)[-1]
+            or "all-MiniLM-L6-v2.gguf2.f16.gguf"
+        )
 
-        span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}"
+        span_name = (
+            f"{SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}"
+        )
 
         with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
             start_time = time.time()
@@ -178,7 +216,7 @@ def embed(version, environment, application_name,
                     capture_message_content=capture_message_content,
                     disable_metrics=disable_metrics,
                     version=version,
-                    **kwargs
+                    **kwargs,
                 )
 
             except Exception as e:

openlit/instrumentation/gpt4all/utils.py

@@ -1,9 +1,14 @@
 """
 GPT4All 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 (
@@ -16,12 +21,14 @@ from openlit.__helpers import (
 )
 from openlit.semcov import SemanticConvention
 
+
 def format_content(prompt):
     """
     Process a prompt to extract content.
     """
     return str(prompt) if prompt else ""
 
+
 def process_chunk(scope, chunk):
     """
     Process a chunk of response data and update state.
@@ -38,8 +45,22 @@ def process_chunk(scope, chunk):
     scope._llmresponse += chunk
     scope._end_time = time.time()
 
-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 generate and embed operations.
     """
@@ -58,9 +79,25 @@ def common_span_attributes(scope, gen_ai_operation, gen_ai_system, server_addres
     scope._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TTFT, ttft)
     scope._span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION, version)
 
-def record_completion_metrics(metrics, gen_ai_operation, gen_ai_system, server_address, server_port,
-    request_model, response_model, environment, application_name, start_time, end_time,
-    input_tokens, output_tokens, cost, tbt=None, ttft=None):
+
+def record_completion_metrics(
+    metrics,
+    gen_ai_operation,
+    gen_ai_system,
+    server_address,
+    server_port,
+    request_model,
+    response_model,
+    environment,
+    application_name,
+    start_time,
+    end_time,
+    input_tokens,
+    output_tokens,
+    cost,
+    tbt=None,
+    ttft=None,
+):
     """
     Record completion-specific metrics for the operation.
     """
@@ -79,16 +116,31 @@ def record_completion_metrics(metrics, gen_ai_operation, gen_ai_system, server_a
     metrics["genai_requests"].add(1, attributes)
     metrics["genai_prompt_tokens"].add(input_tokens, attributes)
     metrics["genai_completion_tokens"].add(output_tokens, attributes)
-    metrics["genai_client_usage_tokens"].record(input_tokens + output_tokens, attributes)
+    metrics["genai_client_usage_tokens"].record(
+        input_tokens + output_tokens, attributes
+    )
     metrics["genai_cost"].record(cost, attributes)
     if tbt is not None:
         metrics["genai_server_tbt"].record(tbt, attributes)
     if ttft is not None:
         metrics["genai_server_ttft"].record(ttft, attributes)
 
-def record_embedding_metrics(metrics, gen_ai_operation, gen_ai_system, server_address, server_port,
-    request_model, response_model, environment, application_name, start_time, end_time,
-    input_tokens, cost):
+
+def record_embedding_metrics(
+    metrics,
+    gen_ai_operation,
+    gen_ai_system,
+    server_address,
+    server_port,
+    request_model,
+    response_model,
+    environment,
+    application_name,
+    start_time,
+    end_time,
+    input_tokens,
+    cost,
+):
     """
     Record embedding-specific metrics for the operation.
     """
@@ -109,8 +161,18 @@ def record_embedding_metrics(metrics, gen_ai_operation, gen_ai_system, server_ad
     metrics["genai_prompt_tokens"].add(input_tokens, attributes)
     metrics["genai_cost"].record(cost, attributes)
 
-def common_t2s_logic(scope, pricing_info, environment, application_name, metrics,
-    capture_message_content, disable_metrics, version, is_stream):
+
+def common_t2s_logic(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content,
+    disable_metrics,
+    version,
+    is_stream,
+):
     """
     Process generate request and generate Telemetry
     """
@@ -118,7 +180,9 @@ def common_t2s_logic(scope, pricing_info, environment, application_name, metrics
     if len(scope._timestamps) > 1:
         scope._tbt = calculate_tbt(scope._timestamps)
 
-    prompt = format_content(scope._kwargs.get("prompt") or (scope._args[0] if scope._args else "") or "")
+    prompt = format_content(
+        scope._kwargs.get("prompt") or (scope._args[0] if scope._args else "") or ""
+    )
     request_model = scope._request_model
 
     # Calculate tokens using input prompt and aggregated response
@@ -128,36 +192,80 @@ def common_t2s_logic(scope, pricing_info, environment, application_name, metrics
     cost = get_chat_model_cost(request_model, pricing_info, input_tokens, output_tokens)
 
     # Common Span Attributes
-    common_span_attributes(scope,
-        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
-        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_CHAT,
+        SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
+        scope._server_address,
+        scope._server_port,
+        request_model,
+        request_model,
+        environment,
+        application_name,
+        is_stream,
+        scope._tbt,
+        scope._ttft,
+        version,
+    )
 
     # Span Attributes for Request parameters
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY, scope._kwargs.get("repeat_penalty", 1.18))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, scope._kwargs.get("max_tokens", 200))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY, scope._kwargs.get("presence_penalty", 0.0))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, scope._kwargs.get("temp", 0.7))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P, scope._kwargs.get("top_p", 0.4))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_K, scope._kwargs.get("top_k", 40))
-    scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE, "text" if isinstance(scope._llmresponse, str) else "json")
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+        scope._kwargs.get("repeat_penalty", 1.18),
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS,
+        scope._kwargs.get("max_tokens", 200),
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY,
+        scope._kwargs.get("presence_penalty", 0.0),
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_TEMPERATURE, scope._kwargs.get("temp", 0.7)
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_TOP_P, scope._kwargs.get("top_p", 0.4)
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_REQUEST_TOP_K, scope._kwargs.get("top_k", 40)
+    )
+    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, input_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, output_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE, input_tokens + output_tokens)
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, input_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, output_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE, input_tokens + output_tokens
+    )
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
 
     # Span Attributes for Tools
     if scope._tools:
-        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_NAME, scope._tools.get("function","")).get("name","")
-        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_CALL_ID, str(scope._tools.get("id","")))
-        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_ARGS, str(scope._tools.get("function","").get("arguments","")))
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_TOOL_NAME, scope._tools.get("function", "")
+        ).get("name", "")
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_TOOL_CALL_ID, str(scope._tools.get("id", ""))
+        )
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_TOOL_ARGS,
+            str(scope._tools.get("function", "").get("arguments", "")),
+        )
 
     # 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._llmresponse)
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._llmresponse
+        )
 
         # To be removed one the change to span_attributes (from span events) is complete
         scope._span.add_event(
@@ -177,13 +285,36 @@ def common_t2s_logic(scope, pricing_info, environment, application_name, metrics
 
     # Metrics
     if not disable_metrics:
-        record_completion_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
-            scope._server_address, scope._server_port, request_model, request_model, environment,
-            application_name, scope._start_time, scope._end_time, input_tokens, output_tokens,
-            cost, scope._tbt, scope._ttft)
+        record_completion_metrics(
+            metrics,
+            SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+            SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
+            scope._server_address,
+            scope._server_port,
+            request_model,
+            request_model,
+            environment,
+            application_name,
+            scope._start_time,
+            scope._end_time,
+            input_tokens,
+            output_tokens,
+            cost,
+            scope._tbt,
+            scope._ttft,
+        )
 
-def common_embedding_logic(scope, pricing_info, environment, application_name, metrics,
-    capture_message_content, disable_metrics, version):
+
+def common_embedding_logic(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content,
+    disable_metrics,
+    version,
+):
     """
     Process embedding request and generate Telemetry
     """
@@ -196,14 +327,29 @@ def common_embedding_logic(scope, pricing_info, environment, application_name, m
     cost = get_embed_model_cost(request_model, pricing_info, input_tokens)
 
     # Common Span Attributes
-    common_span_attributes(scope,
-        SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING, SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
-        scope._server_address, scope._server_port, request_model, request_model,
-        environment, application_name, False, scope._tbt, scope._ttft, version)
+    common_span_attributes(
+        scope,
+        SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING,
+        SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
+        scope._server_address,
+        scope._server_port,
+        request_model,
+        request_model,
+        environment,
+        application_name,
+        False,
+        scope._tbt,
+        scope._ttft,
+        version,
+    )
 
     # Embedding-specific span attributes
-    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, input_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE, input_tokens)
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, input_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE, input_tokens
+    )
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
 
     # Span Attributes for Content
@@ -211,7 +357,9 @@ def common_embedding_logic(scope, pricing_info, environment, application_name, m
         scope._span.add_event(
             name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
             attributes={
-                SemanticConvention.GEN_AI_CONTENT_PROMPT: str(scope._kwargs.get("input", "")),
+                SemanticConvention.GEN_AI_CONTENT_PROMPT: str(
+                    scope._kwargs.get("input", "")
+                ),
             },
         )
 
@@ -219,21 +367,66 @@ def common_embedding_logic(scope, pricing_info, environment, application_name, m
 
     # Metrics
     if not disable_metrics:
-        record_embedding_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING, SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
-            scope._server_address, scope._server_port, request_model, request_model, environment,
-            application_name, scope._start_time, scope._end_time, input_tokens, cost)
+        record_embedding_metrics(
+            metrics,
+            SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING,
+            SemanticConvention.GEN_AI_SYSTEM_GPT4ALL,
+            scope._server_address,
+            scope._server_port,
+            request_model,
+            request_model,
+            environment,
+            application_name,
+            scope._start_time,
+            scope._end_time,
+            input_tokens,
+            cost,
+        )
 
-def process_streaming_generate_response(scope, pricing_info, environment, application_name, metrics,
-    capture_message_content=False, disable_metrics=False, version=""):
+
+def process_streaming_generate_response(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="",
+):
     """
     Process generate request and generate Telemetry
     """
-    common_t2s_logic(scope, pricing_info, environment, application_name, metrics,
-        capture_message_content, disable_metrics, version, is_stream=True)
+    common_t2s_logic(
+        scope,
+        pricing_info,
+        environment,
+        application_name,
+        metrics,
+        capture_message_content,
+        disable_metrics,
+        version,
+        is_stream=True,
+    )
 
-def process_generate_response(response, request_model, pricing_info, server_port, server_address,
-    environment, application_name, metrics, start_time, span, args, kwargs, capture_message_content=False,
-    disable_metrics=False, version="1.0.0"):
+
+def process_generate_response(
+    response,
+    request_model,
+    pricing_info,
+    server_port,
+    server_address,
+    environment,
+    application_name,
+    metrics,
+    start_time,
+    span,
+    args,
+    kwargs,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="1.0.0",
+):
     """
     Process generate request and generate Telemetry
     """
@@ -252,14 +445,37 @@ def process_generate_response(response, request_model, pricing_info, server_port
     scope._args = args
     scope._tools = None
 
-    common_t2s_logic(scope, pricing_info, environment, application_name, metrics,
-        capture_message_content, disable_metrics, version, is_stream=False)
+    common_t2s_logic(
+        scope,
+        pricing_info,
+        environment,
+        application_name,
+        metrics,
+        capture_message_content,
+        disable_metrics,
+        version,
+        is_stream=False,
+    )
 
     return response
 
-def process_embedding_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):
+
+def process_embedding_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 embedding request and generate Telemetry
     """
@@ -275,7 +491,15 @@ def process_embedding_response(response, request_model, pricing_info, server_por
     scope._server_address, scope._server_port = server_address, server_port
     scope._kwargs = kwargs
 
-    common_embedding_logic(scope, pricing_info, environment, application_name, metrics,
-        capture_message_content, disable_metrics, version)
+    common_embedding_logic(
+        scope,
+        pricing_info,
+        environment,
+        application_name,
+        metrics,
+        capture_message_content,
+        disable_metrics,
+        version,
+    )
 
     return response