openlit 1.34.16__py3-none-any.whl → 1.34.18__py3-none-any.whl

openlit/instrumentation/mistral/utils.py

@@ -0,0 +1,298 @@
+"""
+Mistral 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,
+    get_embed_model_cost,
+    common_span_attributes,
+    record_completion_metrics,
+    record_embedding_metrics,
+)
+from openlit.semcov import SemanticConvention
+
+def format_content(messages):
+    """
+    Process a list of messages to extract content.
+    """
+
+    formatted_messages = []
+    for message in messages:
+        role = message.get("role", "user")
+        content = message.get("content", "")
+
+        if isinstance(content, list):
+            content_str = ", ".join(
+                f'{item["type"]}: {item["text"] if "text" in item else item.get("image_url", "")}'
+                if "type" in item else f'text: {item.get("text", "")}'
+                for item in content
+            )
+            formatted_messages.append(f"{role}: {content_str}")
+        else:
+            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 aggregated response from events
+    if chunked.get("data"):
+        data = chunked.get("data")
+        choices = data.get("choices", [])
+
+        if choices and "delta" in choices[0]:
+            delta = choices[0]["delta"]
+            content = delta.get("content")
+            if content:
+                scope._llmresponse += content
+
+            # Handle tool calls in streaming - optimized
+            delta_tools = delta.get("tool_calls")
+            if delta_tools:
+                scope._tools = scope._tools or []
+
+                for tool in delta_tools:
+                    idx = tool.get("index", 0)
+
+                    # Extend list if needed
+                    scope._tools.extend([{}] * (idx + 1 - len(scope._tools)))
+
+                    if tool.get("id"):  # New tool (id exists)
+                        func = tool.get("function", {})
+                        scope._tools[idx] = {
+                            "id": tool["id"],
+                            "function": {"name": func.get("name", ""), "arguments": func.get("arguments", "")},
+                            "type": tool.get("type", "function")
+                        }
+                    elif scope._tools[idx] and "function" in tool:  # Append args (id is None)
+                        scope._tools[idx]["function"]["arguments"] += tool["function"].get("arguments", "")
+
+        # Handle usage information (typically only in final chunk)
+        if data.get("usage"):
+            scope._input_tokens = data.get("usage").get("prompt_tokens", 0)
+            scope._output_tokens = data.get("usage").get("completion_tokens", 0)
+            scope._response_id = data.get("id")
+            scope._response_model = data.get("model")
+            scope._finish_reason = choices[0].get("finish_reason", "") if choices else ""
+            scope._end_time = time.time()
+
+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
+    """
+
+    if len(scope._timestamps) > 1:
+        scope._tbt = calculate_tbt(scope._timestamps)
+
+    prompt = format_content(scope._kwargs.get("messages", []))
+    request_model = scope._kwargs.get("model", "mistral-small-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_MISTRAL,
+        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_SEED, scope._kwargs.get("seed", ""))
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_FREQUENCY_PENALTY, scope._kwargs.get("frequency_penalty", 0.0))
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MAX_TOKENS, scope._kwargs.get("max_tokens", -1))
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_PRESENCE_PENALTY, scope._kwargs.get("presence_penalty", 0.0))
+    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", 0.3))
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_K, scope._kwargs.get("k", 1.0))
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_TOP_P, scope._kwargs.get("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)
+
+    # Span Attributes for Tools - optimized
+    if scope._tools:
+        tools = scope._tools if isinstance(scope._tools, list) else [scope._tools]
+
+        names, ids, args = zip(*[
+            (t.get("function", {}).get("name", ""),
+             str(t.get("id", "")),
+             str(t.get("function", {}).get("arguments", "")))
+            for t in tools if isinstance(t, dict) and t
+        ]) if tools else ([], [], [])
+
+        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_NAME, ", ".join(filter(None, names)))
+        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_CALL_ID, ", ".join(filter(None, ids)))
+        scope._span.set_attribute(SemanticConvention.GEN_AI_TOOL_ARGS, ", ".join(filter(None, args)))
+
+    # 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)
+
+        # 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: prompt,
+            },
+        )
+        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))
+
+    # Metrics
+    if not disable_metrics:
+        record_completion_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_MISTRAL,
+            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 request and generate Telemetry
+    """
+
+    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 chat request and generate Telemetry
+    """
+
+    # Create scope object
+    scope = type("GenericScope", (), {})()
+    response_dict = response_as_dict(response)
+
+    scope._start_time = start_time
+    scope._end_time = time.time()
+    scope._span = span
+    scope._llmresponse = " ".join(
+        (choice.get("message", {}).get("content") or "")
+        for choice in response_dict.get("choices", [])
+    )
+    scope._response_id = response_dict.get("id")
+    scope._response_model = response_dict.get("model")
+    scope._input_tokens = response_dict.get("usage", {}).get("prompt_tokens", 0)
+    scope._output_tokens = response_dict.get("usage", {}).get("completion_tokens", 0)
+    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
+    scope._finish_reason = str(response_dict.get("choices", [])[0].get("finish_reason", "")) if response_dict.get("choices") else ""
+
+    # Handle tool calls
+    if kwargs.get("tools"):
+        scope._tools = response_dict.get("choices", [{}])[0].get("message", {}).get("tool_calls")
+    else:
+        scope._tools = None
+
+    common_chat_logic(scope, pricing_info, environment, application_name, metrics,
+        capture_message_content, disable_metrics, version, is_stream=False)
+
+    return response
+
+def common_embedding_logic(scope, pricing_info, environment, application_name, metrics,
+    capture_message_content, disable_metrics, version):
+    """
+    Process embedding request and generate Telemetry
+    """
+
+    request_model = scope._kwargs.get("model", "mistral-embed")
+    inputs = scope._kwargs.get("inputs", [])
+
+    cost = get_embed_model_cost(request_model, pricing_info, scope._input_tokens)
+
+    # Common Span Attributes
+    common_span_attributes(scope,
+        SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING, SemanticConvention.GEN_AI_SYSTEM_MISTRAL,
+        scope._server_address, scope._server_port, request_model, scope._response_model,
+        environment, application_name, False, 0, scope._ttft, version)
+
+    # Span Attributes for Request parameters
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_ENCODING_FORMATS, [scope._kwargs.get("encoding_format", "float")])
+
+    # 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_CLIENT_TOKEN_USAGE, scope._input_tokens)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
+
+    # Span Attributes for Content
+    if capture_message_content:
+        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, str(inputs))
+
+        # 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: str(inputs),
+            },
+        )
+
+    scope._span.set_status(Status(StatusCode.OK))
+
+    # Metrics
+    if not disable_metrics:
+        record_embedding_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING, SemanticConvention.GEN_AI_SYSTEM_MISTRAL,
+            scope._server_address, scope._server_port, request_model, scope._response_model, environment,
+            application_name, scope._start_time, scope._end_time, scope._input_tokens, cost)
+
+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
+    """
+
+    # Create scope object
+    scope = type("GenericScope", (), {})()
+    response_dict = response_as_dict(response)
+
+    scope._start_time = start_time
+    scope._end_time = time.time()
+    scope._span = span
+    scope._response_model = response_dict.get("model")
+    scope._input_tokens = response_dict.get("usage", {}).get("prompt_tokens", 0)
+    scope._ttft = scope._end_time - scope._start_time
+    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)
+
+    return response

{openlit-1.34.16.dist-info → openlit-1.34.18.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: openlit
-Version: 1.34.16
+Version: 1.34.18
 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-1.34.16.dist-info → openlit-1.34.18.dist-info}/RECORD

@@ -38,9 +38,10 @@ openlit/instrumentation/bedrock/bedrock.py,sha256=kP9ESKzqhWu-dIWseyaeyendUyo6b7
 openlit/instrumentation/bedrock/utils.py,sha256=_mTUIbioEg4jfoxocUbfc7RgGjhm9ACelbxIoFu4jbM,11636
 openlit/instrumentation/chroma/__init__.py,sha256=4ZeHY1OInRKQbb4qg8BVvGJtWN1XdzW6mosqi7-6ruE,3353
 openlit/instrumentation/chroma/chroma.py,sha256=fPIX6vgcDFkBgKGxrkXZWDaBtZa5y-PV3jJjKsFqjDg,10640
-openlit/instrumentation/cohere/__init__.py,sha256=TIRq1obu-zqBji0HhMbFGfI2q5m-zw0nWbToKeZqpg4,2905
-openlit/instrumentation/cohere/async_cohere.py,sha256=cFIrQ-maKmM8JKRitX-mgD7Jm_1UDyO_kOp4X_0MQww,30745
-openlit/instrumentation/cohere/cohere.py,sha256=11QGsBTLiTfF-oH_fIQ9IBbIgtS2vf7M-VvbG8MuO_Q,30628
+openlit/instrumentation/cohere/__init__.py,sha256=FIJ_QokUZcN9UsVPWckRfHiLVJ5zj3jtcOjQmjQA5f0,2978
+openlit/instrumentation/cohere/async_cohere.py,sha256=wgctJa-BcKEkVguLorR7mYRShv7GiioF1_zxevvzXTw,6990
+openlit/instrumentation/cohere/cohere.py,sha256=7YAj6kwJ1MDLFY3WtbPRehZFIseDVIbvPQJZxG8Qc6A,6814
+openlit/instrumentation/cohere/utils.py,sha256=5l3Av-t0GzawPd0j5L9jcpdJDGOvLq2YCMpo3usS4OY,14533
 openlit/instrumentation/controlflow/__init__.py,sha256=Y5xSj6bwkGufG989YtRGujzj8qrV4T5kXr4hzGb2nGk,2168
 openlit/instrumentation/controlflow/controlflow.py,sha256=NNovai9DwRCSaa-shFxAyUIYiRaEBgXkyuXFk3lQlPU,5573
 openlit/instrumentation/crawl4ai/__init__.py,sha256=TTFFeZkTgPZJAD1Tlr2LK-CTH2A7NkZK-5SHKFfFOtU,1947
@@ -90,9 +91,10 @@ openlit/instrumentation/mem0/__init__.py,sha256=IadP3bKgz2HCbnrh9S7AW24uDauGkzsI
 openlit/instrumentation/mem0/mem0.py,sha256=Y3Y_OWH9fW4vrus5vwPoWyy7sWXB9IR3k5tt6idMJ9E,5348
 openlit/instrumentation/milvus/__init__.py,sha256=qQqI-mOXEIsXGYwLyUGLEECcIHCm_6_KQZq0vFL90LY,2991
 openlit/instrumentation/milvus/milvus.py,sha256=G_wUAvwKmm7Jghd-X8xdwZphRjnJSKctU1424pQmJ3U,9220
-openlit/instrumentation/mistral/__init__.py,sha256=5hh5s2-kudMbfkWD6Lhe5IQvrlZ5qA-Sv3laeXwyaPA,3060
-openlit/instrumentation/mistral/async_mistral.py,sha256=XgT7ZQ2Dx_UscoljjgKvWb6Fe5FuDs7Yrwau96KpSKQ,31187
-openlit/instrumentation/mistral/mistral.py,sha256=_2qM8v4RCL-S0Mm1vbW77m5vUm8aPDyiijbP75yHZaY,31083
+openlit/instrumentation/mistral/__init__.py,sha256=D4CLrx9KSSxAPA1m00743Og0Tl8BS47nsgp4qG5-qh8,2977
+openlit/instrumentation/mistral/async_mistral.py,sha256=LWq8tYahbA7NOPDdk6DWJAKuxR4GHOC49w6L1QuNtBw,6946
+openlit/instrumentation/mistral/mistral.py,sha256=lyXyPZuxVACcZoOz85G4FFH4KP77uGf3aOiFDdHLQFI,6771
+openlit/instrumentation/mistral/utils.py,sha256=B_sdUXXYPq6w3-qPX6lei0eKU6OxcgXOKG40TOZWDUQ,13200
 openlit/instrumentation/multion/__init__.py,sha256=Wr3lcDyG_YbOLkCUzBFhraAedF6E113tce8eSWlcz10,3149
 openlit/instrumentation/multion/async_multion.py,sha256=XutZnayCJOZ_NA9bvE1NUoej41KOGR7FRn2tpoGKMEU,6092
 openlit/instrumentation/multion/multion.py,sha256=-WqRAcu5qiEMY9XDmlJTQHuQiWfdwms9JDn127QCNb8,6074
@@ -139,7 +141,7 @@ openlit/otel/events.py,sha256=VrMjTpvnLtYRBHCiFwJojTQqqNpRCxoD4yJYeQrtPsk,3560
 openlit/otel/metrics.py,sha256=GM2PDloBGRhBTkHHkYaqmOwIAQkY124ZhW4sEqW1Fgk,7086
 openlit/otel/tracing.py,sha256=tjV2bEbEDPUB1Z46gE-UsJsb04sRdFrfbhIDkxViZc0,3103
 openlit/semcov/__init__.py,sha256=8oIh2VC667NDh8FA3M-ESusHmeus1sgDUD8binx_nAc,13519
-openlit-1.34.16.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-openlit-1.34.16.dist-info/METADATA,sha256=mdKEEE4FgRuUOe_Pl1Crh0S89-A8wCrBchSFJ7cqBRI,23470
-openlit-1.34.16.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-openlit-1.34.16.dist-info/RECORD,,
+openlit-1.34.18.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+openlit-1.34.18.dist-info/METADATA,sha256=H34xsI30squSeUxI--7xhbuKEcOc_fEl7ru4HsnOJZU,23470
+openlit-1.34.18.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+openlit-1.34.18.dist-info/RECORD,,