openlit 1.18.0__py3-none-any.whl → 1.18.2__py3-none-any.whl

openlit/instrumentation/mistral/__init__.py

@@ -9,10 +9,10 @@ from openlit.instrumentation.mistral.mistral import chat, chat_stream, embedding
 from openlit.instrumentation.mistral.async_mistral import async_chat, async_chat_stream
 from openlit.instrumentation.mistral.async_mistral import async_embeddings
 
-_instruments = ("mistralai >= 0.1.0",)
+_instruments = ("mistralai >= 1.0.0",)
 
 class MistralInstrumentor(BaseInstrumentor):
-    """An instrumentor for Azure Mistral's client library."""
+    """An instrumentor for Mistral's client library."""
 
     def instrumentation_dependencies(self) -> Collection[str]:
         return _instruments
@@ -27,50 +27,50 @@ class MistralInstrumentor(BaseInstrumentor):
         disable_metrics = kwargs.get("disable_metrics")
         version = importlib.metadata.version("mistralai")
 
-        #sync
+        # sync
         wrap_function_wrapper(
-            "mistralai.client",  
-            "MistralClient.chat",  
+            "mistralai.chat",  
+            "Chat.complete",  
             chat("mistral.chat", version, environment, application_name,
                  tracer, pricing_info, trace_content, metrics, disable_metrics),
         )
 
-        #sync
+        # sync
         wrap_function_wrapper(
-            "mistralai.client",  
-            "MistralClient.chat_stream",  
+            "mistralai.chat",  
+            "Chat.stream",  
             chat_stream("mistral.chat", version, environment, application_name,
                         tracer, pricing_info, trace_content, metrics, disable_metrics),
         )
 
-        #sync
+        # sync
         wrap_function_wrapper(
-            "mistralai.client",  
-            "MistralClient.embeddings",  
+            "mistralai.embeddings",  
+            "Embeddings.create",  
             embeddings("mistral.embeddings", version, environment, application_name,
                        tracer, pricing_info, trace_content, metrics, disable_metrics),
         )
 
         # Async
         wrap_function_wrapper(
-            "mistralai.async_client",  
-            "MistralAsyncClient.chat",  
+            "mistralai.chat",  
+            "Chat.complete_async",  
             async_chat("mistral.chat", version, environment, application_name,
                        tracer, pricing_info, trace_content, metrics, disable_metrics),
         )
 
-        #sync
+        # Async
         wrap_function_wrapper(
-            "mistralai.async_client",  
-            "MistralAsyncClient.chat_stream",  
+            "mistralai.chat",  
+            "Chat.stream_async",  
             async_chat_stream("mistral.chat", version, environment, application_name,
                               tracer, pricing_info, trace_content, metrics, disable_metrics),
         )
 
         #sync
         wrap_function_wrapper(
-            "mistralai.async_client",  
-            "MistralAsyncClient.embeddings",  
+            "mistralai.embeddings",  
+            "Embeddings.create_async",  
             async_embeddings("mistral.embeddings", version, environment, application_name,
                              tracer, pricing_info, trace_content, metrics, disable_metrics),
         )

openlit/instrumentation/mistral/async_mistral.py

@@ -56,8 +56,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
                 message_prompt = kwargs.get('messages', "")
                 formatted_messages = []
                 for message in message_prompt:
-                    role = message.role
-                    content = message.content
+                    role = message["role"]
+                    content = message["content"]
 
                     if isinstance(content, list):
                         content_str = ", ".join(
@@ -207,14 +207,14 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
                 llmresponse = ""
 
                 # Loop through streaming events capturing relevant details
-                async for event in wrapped(*args, **kwargs):
-                    response_id = event.id
-                    llmresponse += event.choices[0].delta.content
-                    if event.usage is not None:
-                        prompt_tokens = event.usage.prompt_tokens
-                        completion_tokens = event.usage.completion_tokens
-                        total_tokens = event.usage.total_tokens
-                        finish_reason = event.choices[0].finish_reason
+                async for event in await wrapped(*args, **kwargs):
+                    response_id = event.data.id
+                    llmresponse += event.data.choices[0].delta.content
+                    if event.data.usage is not None:
+                        prompt_tokens = event.data.usage.prompt_tokens
+                        completion_tokens = event.data.usage.completion_tokens
+                        total_tokens = event.data.usage.total_tokens
+                        finish_reason = event.data.choices[0].finish_reason
                     yield event
 
                 # Handling exception ensure observability without disrupting operation
@@ -223,8 +223,8 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
                     message_prompt = kwargs.get('messages', "")
                     formatted_messages = []
                     for message in message_prompt:
-                        role = message.role
-                        content = message.content
+                        role = message["role"]
+                        content = message["content"]
 
                         if isinstance(content, list):
                             content_str = ", ".join(
@@ -364,7 +364,7 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
 
             try:
                 # Get prompt from kwargs and store as a single string
-                prompt = ', '.join(kwargs.get('input', []))
+                prompt = ', '.join(kwargs.get('inputs', []))
 
                 # Calculate cost of the operation
                 cost = get_embed_model_cost(kwargs.get('model', "mistral-embed"),

openlit/instrumentation/mistral/mistral.py

@@ -55,8 +55,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
                 message_prompt = kwargs.get('messages', "")
                 formatted_messages = []
                 for message in message_prompt:
-                    role = message.role
-                    content = message.content
+                    role = message["role"]
+                    content = message["content"]
 
                     if isinstance(content, list):
                         content_str = ", ".join(
@@ -207,13 +207,13 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
 
                 # Loop through streaming events capturing relevant details
                 for event in wrapped(*args, **kwargs):
-                    response_id = event.id
-                    llmresponse += event.choices[0].delta.content
-                    if event.usage is not None:
-                        prompt_tokens = event.usage.prompt_tokens
-                        completion_tokens = event.usage.completion_tokens
-                        total_tokens = event.usage.total_tokens
-                        finish_reason = event.choices[0].finish_reason
+                    response_id = event.data.id
+                    llmresponse += event.data.choices[0].delta.content
+                    if event.data.usage is not None:
+                        prompt_tokens = event.data.usage.prompt_tokens
+                        completion_tokens = event.data.usage.completion_tokens
+                        total_tokens = event.data.usage.total_tokens
+                        finish_reason = event.data.choices[0].finish_reason
                     yield event
 
                 # Handling exception ensure observability without disrupting operation
@@ -222,8 +222,8 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
                     message_prompt = kwargs.get('messages', "")
                     formatted_messages = []
                     for message in message_prompt:
-                        role = message.role
-                        content = message.content
+                        role = message["role"]
+                        content = message["content"]
 
                         if isinstance(content, list):
                             content_str = ", ".join(
@@ -363,7 +363,7 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
 
             try:
                 # Get prompt from kwargs and store as a single string
-                prompt = ', '.join(kwargs.get('input', []))
+                prompt = ', '.join(kwargs.get('inputs', []))
 
                 # Calculate cost of the operation
                 cost = get_embed_model_cost(kwargs.get('model', "mistral-embed"),

openlit/instrumentation/ollama/async_ollama.py

@@ -6,7 +6,11 @@ Module for monitoring Ollama API calls.
 import logging
 from opentelemetry.trace import SpanKind, Status, StatusCode
 from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
-from openlit.__helpers import handle_exception, general_tokens
+from openlit.__helpers import (
+    handle_exception,
+    general_tokens,
+    get_chat_model_cost,
+    get_embed_model_cost)
 from openlit.semcov import SemanticConvetion
 
 # Initialize logger for logging potential issues and operations
@@ -90,10 +94,11 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
                                 formatted_messages.append(f"{role}: {content}")
                         prompt = "\n".join(formatted_messages)
 
-                        # Calculate cost of the operation
-                        cost = 0
                         prompt_tokens = general_tokens(prompt)
                         total_tokens = prompt_tokens + completion_tokens
+                        # Calculate cost of the operation
+                        cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                         # Set Span attributes
                         span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
@@ -219,11 +224,12 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
                             },
                         )
 
-                    # Calculate cost of the operation
-                    cost = 0
                     prompt_tokens = general_tokens(prompt)
                     completion_tokens = response["eval_count"]
                     total_tokens = prompt_tokens + completion_tokens
+                    # Calculate cost of the operation
+                    cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                     span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
                                         prompt_tokens)
@@ -331,10 +337,11 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
 
                     # Handling exception ensure observability without disrupting operation
                     try:
-                        # Calculate cost of the operation
-                        cost = 0
                         prompt_tokens = general_tokens(kwargs.get("prompt", ""))
                         total_tokens = prompt_tokens + completion_tokens
+                        # Calculate cost of the operation
+                        cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                         # Set Span attributes
                         span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
@@ -442,11 +449,12 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
                             },
                         )
 
-                    # Calculate cost of the operation
-                    cost = 0
                     prompt_tokens = response["prompt_eval_count"]
                     completion_tokens = response["eval_count"]
                     total_tokens = prompt_tokens + completion_tokens
+                    # Calculate cost of the operation
+                    cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                     span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
                                         prompt_tokens)
@@ -534,9 +542,10 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
             response = await wrapped(*args, **kwargs)
 
             try:
-                # Calculate cost of the operation
-                cost = 0
                 prompt_tokens = general_tokens(kwargs.get('prompt', ""))
+                # Calculate cost of the operation
+                cost = get_embed_model_cost(kwargs.get('model', "mistral-embed"),
+                                            pricing_info, prompt_tokens)
                 # Set Span attributes
                 span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
                 span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,

openlit/instrumentation/ollama/ollama.py

@@ -6,7 +6,12 @@ Module for monitoring Ollama API calls.
 import logging
 from opentelemetry.trace import SpanKind, Status, StatusCode
 from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
-from openlit.__helpers import handle_exception, general_tokens
+from openlit.__helpers import (
+    handle_exception,
+    general_tokens,
+    get_chat_model_cost,
+    get_embed_model_cost
+)
 from openlit.semcov import SemanticConvetion
 
 # Initialize logger for logging potential issues and operations
@@ -90,10 +95,11 @@ def chat(gen_ai_endpoint, version, environment, application_name,
                                 formatted_messages.append(f"{role}: {content}")
                         prompt = "\n".join(formatted_messages)
 
-                        # Calculate cost of the operation
-                        cost = 0
                         prompt_tokens = general_tokens(prompt)
                         total_tokens = prompt_tokens + completion_tokens
+                        # Calculate cost of the operation
+                        cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                         # Set Span attributes
                         span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
@@ -219,11 +225,12 @@ def chat(gen_ai_endpoint, version, environment, application_name,
                             },
                         )
 
-                    # Calculate cost of the operation
-                    cost = 0
                     prompt_tokens = general_tokens(prompt)
                     completion_tokens = response["eval_count"]
                     total_tokens = prompt_tokens + completion_tokens
+                    # Calculate cost of the operation
+                    cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                     span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
                                         prompt_tokens)
@@ -331,10 +338,11 @@ def generate(gen_ai_endpoint, version, environment, application_name,
 
                     # Handling exception ensure observability without disrupting operation
                     try:
-                        # Calculate cost of the operation
-                        cost = 0
                         prompt_tokens = general_tokens(kwargs.get("prompt", ""))
                         total_tokens = prompt_tokens + completion_tokens
+                        # Calculate cost of the operation
+                        cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                         # Set Span attributes
                         span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
@@ -442,11 +450,12 @@ def generate(gen_ai_endpoint, version, environment, application_name,
                             },
                         )
 
-                    # Calculate cost of the operation
-                    cost = 0
                     prompt_tokens = response["prompt_eval_count"]
                     completion_tokens = response["eval_count"]
                     total_tokens = prompt_tokens + completion_tokens
+                    # Calculate cost of the operation
+                    cost = get_chat_model_cost(kwargs.get("model", "llama3"),
+                                                pricing_info, prompt_tokens, completion_tokens)
 
                     span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
                                         prompt_tokens)
@@ -534,9 +543,10 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
             response = wrapped(*args, **kwargs)
 
             try:
-                # Calculate cost of the operation
-                cost = 0
                 prompt_tokens = general_tokens(kwargs.get('prompt', ""))
+                # Calculate cost of the operation
+                cost = get_embed_model_cost(kwargs.get('model', "mistral-embed"),
+                                            pricing_info, prompt_tokens)
                 # Set Span attributes
                 span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
                 span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,

{openlit-1.18.0.dist-info → openlit-1.18.2.dist-info}/METADATA

@@ -1,9 +1,9 @@
 Metadata-Version: 2.1
 Name: openlit
-Version: 1.18.0
-Summary: OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications, facilitating the integration of observability into your GenAI-driven projects
+Version: 1.18.2
+Summary: OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects
 Home-page: https://github.com/openlit/openlit/tree/main/openlit/python
-Keywords: OpenTelemetry,otel,otlp,llm,tracing,openai,anthropic,claude,cohere,llm monitoring,observability,monitoring,gpt,Generative AI,chatGPT
+Keywords: OpenTelemetry,otel,otlp,llm,tracing,openai,anthropic,claude,cohere,llm monitoring,observability,monitoring,gpt,Generative AI,chatGPT,gpu
 Author: OpenLIT
 Requires-Python: >=3.7.1,<4.0.0
 Classifier: Programming Language :: Python :: 3

{openlit-1.18.0.dist-info → openlit-1.18.2.dist-info}/RECORD

@@ -28,12 +28,12 @@ openlit/instrumentation/llamaindex/__init__.py,sha256=vPtK65G6b-TwJERowVRUVl7f_n
 openlit/instrumentation/llamaindex/llamaindex.py,sha256=uiIigbwhonSbJWA7LpgOVI1R4kxxPODS1K5wyHIQ4hM,4048
 openlit/instrumentation/milvus/__init__.py,sha256=qi1yfmMrvkDtnrN_6toW8qC9BRL78bq7ayWpObJ8Bq4,2961
 openlit/instrumentation/milvus/milvus.py,sha256=qhKIoggBAJhRctRrBYz69AcvXH-eh7oBn_l9WfxpAjI,9121
-openlit/instrumentation/mistral/__init__.py,sha256=zJCIpFWRbsYrvooOJYuqwyuKeSOQLWbyXWCObL-Snks,3156
-openlit/instrumentation/mistral/async_mistral.py,sha256=WXU46nbe61IglfGehrwdprMVB6hNiuqqmJHE3XvvP0E,22192
-openlit/instrumentation/mistral/mistral.py,sha256=DC-wLIypokPEEAbVSKX5sytv94DY2QDnk12401e5vq8,22039
+openlit/instrumentation/mistral/__init__.py,sha256=niWn0gYNOTPS5zoTjtCciDqQVj-iJehnpdh7ElB-H9w,3088
+openlit/instrumentation/mistral/async_mistral.py,sha256=l-kcaGPrX3sqPH-RXWo6ope0Ui3nUvExNJ4KX9QgDMY,22246
+openlit/instrumentation/mistral/mistral.py,sha256=Q7MMRvVFsM8o0_ebZ0EfnhGjs16SJSnmu-oE798gYMQ,22087
 openlit/instrumentation/ollama/__init__.py,sha256=cOax8PiypDuo_FC4WvDCYBRo7lH5nV9xU92h7k-eZbg,3812
-openlit/instrumentation/ollama/async_ollama.py,sha256=UBwl-Jv4VPyPuhKIesL0VYIQ2u1AZ0U8c9MZQzEPn6c,30594
-openlit/instrumentation/ollama/ollama.py,sha256=bZPgkJJUD3RgJTUUt98cklfZ8Y1lsSgQGOUXXHk53BI,30504
+openlit/instrumentation/ollama/async_ollama.py,sha256=7lbikD-I9k8VL63idqj3VMEfiEKJmFNUPR8Xb6g2phQ,31366
+openlit/instrumentation/ollama/ollama.py,sha256=lBt1d3rFnF1tFbfdOccwjEafHnmTAUGsiOKSHku6Fkw,31277
 openlit/instrumentation/openai/__init__.py,sha256=AZ2cPr3TMKkgGdMl_yXMeSi7bWhtmMqOW1iHdzHHGHA,16265
 openlit/instrumentation/openai/async_azure_openai.py,sha256=XbST1UE_zXzNL6RX2XwCsK_a6IhG9PHVTMKBjGrUcB0,48961
 openlit/instrumentation/openai/async_openai.py,sha256=RGNpKLsHYfJXjj1ImuWRJToVSs0wdvMNp2kyTBrBaDw,47578
@@ -53,7 +53,7 @@ openlit/instrumentation/vllm/vllm.py,sha256=lDzM7F5pgxvh8nKL0dcKB4TD0Mc9wXOWeXOs
 openlit/otel/metrics.py,sha256=O7NoaDz0bY19mqpE4-0PcKwEe-B-iJFRgOCaanAuZAc,4291
 openlit/otel/tracing.py,sha256=vL1ifMbARPBpqK--yXYsCM6y5dSu5LFIKqkhZXtYmUc,3712
 openlit/semcov/__init__.py,sha256=EvoNOKtc7UKwLZ3Gp0-B1zwmeTcAIbx8O7wvAw8wXP4,7498
-openlit-1.18.0.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-openlit-1.18.0.dist-info/METADATA,sha256=LzRSgLCKr0x6Vr8YwhFkPCEuTbF_NfQtpQTpMdiXCqo,14334
-openlit-1.18.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-openlit-1.18.0.dist-info/RECORD,,
+openlit-1.18.2.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+openlit-1.18.2.dist-info/METADATA,sha256=eD-PsH7RbUA7EOICMoUHzO_f1g_Sa0b6UcNvHnBwY-8,14347
+openlit-1.18.2.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+openlit-1.18.2.dist-info/RECORD,,