openlit 1.14.2__py3-none-any.whl → 1.16.2__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (23) hide show
  1. openlit/__init__.py +3 -0
  2. openlit/instrumentation/anthropic/anthropic.py +2 -2
  3. openlit/instrumentation/anthropic/async_anthropic.py +2 -2
  4. openlit/instrumentation/bedrock/__init__.py +3 -3
  5. openlit/instrumentation/bedrock/bedrock.py +69 -307
  6. openlit/instrumentation/cohere/cohere.py +2 -2
  7. openlit/instrumentation/elevenlabs/__init__.py +70 -0
  8. openlit/instrumentation/elevenlabs/async_elevenlabs.py +117 -0
  9. openlit/instrumentation/elevenlabs/elevenlabs.py +123 -0
  10. openlit/instrumentation/groq/async_groq.py +10 -10
  11. openlit/instrumentation/groq/groq.py +10 -10
  12. openlit/instrumentation/mistral/async_mistral.py +4 -4
  13. openlit/instrumentation/mistral/mistral.py +4 -4
  14. openlit/instrumentation/openai/async_azure_openai.py +12 -12
  15. openlit/instrumentation/openai/async_openai.py +10 -10
  16. openlit/instrumentation/openai/azure_openai.py +12 -12
  17. openlit/instrumentation/openai/openai.py +10 -10
  18. openlit/instrumentation/transformers/transformers.py +1 -1
  19. openlit/semcov/__init__.py +5 -3
  20. {openlit-1.14.2.dist-info → openlit-1.16.2.dist-info}/METADATA +16 -16
  21. {openlit-1.14.2.dist-info → openlit-1.16.2.dist-info}/RECORD +23 -20
  22. {openlit-1.14.2.dist-info → openlit-1.16.2.dist-info}/LICENSE +0 -0
  23. {openlit-1.14.2.dist-info → openlit-1.16.2.dist-info}/WHEEL +0 -0
openlit/instrumentation/elevenlabs/async_elevenlabs.py ADDED
@@ -0,0 +1,117 @@
1
+ # pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, possibly-used-before-assignment
2
+ """
3
+ Module for monitoring Ollama API calls.
4
+ """
5
+
6
+ import logging
7
+ from opentelemetry.trace import SpanKind, Status, StatusCode
8
+ from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
9
+ from openlit.__helpers import get_audio_model_cost
10
+ from openlit.__helpers import handle_exception
11
+ from openlit.semcov import SemanticConvetion
12
+
13
+ # Initialize logger for logging potential issues and operations
14
+ logger = logging.getLogger(__name__)
15
+
16
+ def async_generate(gen_ai_endpoint, version, environment, application_name,
17
+ tracer, pricing_info, trace_content, metrics, disable_metrics):
18
+ """
19
+ Generates a telemetry wrapper for creating speech audio to collect metrics.
20
+
21
+ Args:
22
+ gen_ai_endpoint: Endpoint identifier for logging and tracing.
23
+ version: Version of the monitoring package.
24
+ environment: Deployment environment (e.g., production, staging).
25
+ application_name: Name of the application using the ElevenLabs API.
26
+ tracer: OpenTelemetry tracer for creating spans.
27
+ pricing_info: Information used for calculating the cost of generating speech audio.
28
+ trace_content: Flag indicating whether to trace the input text and generated audio.
29
+
30
+ Returns:
31
+ A function that wraps the speech audio creation method to add telemetry.
32
+ """
33
+
34
+ async def wrapper(wrapped, instance, args, kwargs):
35
+ """
36
+ Wraps the 'generate' API call to add telemetry.
37
+
38
+ This collects metrics such as execution time, cost, and handles errors
39
+ gracefully, adding details to the trace for observability.
40
+
41
+ Args:
42
+ wrapped: The original 'generate' method to be wrapped.
43
+ instance: The instance of the class where the original method is defined.
44
+ args: Positional arguments for the 'generate' method.
45
+ kwargs: Keyword arguments for the 'generate' method.
46
+
47
+ Returns:
48
+ The response from the original 'generate' method.
49
+ """
50
+
51
+ with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
52
+ response = await wrapped(*args, **kwargs)
53
+
54
+ try:
55
+ # Calculate cost of the operation
56
+ cost = get_audio_model_cost(kwargs.get("model", "eleven_multilingual_v2"),
57
+ pricing_info, kwargs.get("text", ""))
58
+
59
+ # Set Span attributes
60
+ span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
61
+ span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
62
+ SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS)
63
+ span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
64
+ SemanticConvetion.GEN_AI_TYPE_AUDIO)
65
+ span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
66
+ gen_ai_endpoint)
67
+ span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
68
+ environment)
69
+ span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
70
+ application_name)
71
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
72
+ kwargs.get("model", "eleven_multilingual_v2"))
73
+ if isinstance(kwargs.get("voice", "Rachel"), str):
74
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_VOICE,
75
+ kwargs.get("voice", "Rachel"))
76
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT,
77
+ kwargs.get("output_format", "mp3"))
78
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_SETTINGS,
79
+ str(kwargs.get("voice_settings", "")))
80
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
81
+ cost)
82
+ if trace_content:
83
+ span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
84
+ str(kwargs.get("text", "")))
85
+
86
+ span.set_status(Status(StatusCode.OK))
87
+
88
+ if disable_metrics is False:
89
+ attributes = {
90
+ TELEMETRY_SDK_NAME:
91
+ "openlit",
92
+ SemanticConvetion.GEN_AI_APPLICATION_NAME:
93
+ application_name,
94
+ SemanticConvetion.GEN_AI_SYSTEM:
95
+ SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS,
96
+ SemanticConvetion.GEN_AI_ENVIRONMENT:
97
+ environment,
98
+ SemanticConvetion.GEN_AI_TYPE:
99
+ SemanticConvetion.GEN_AI_TYPE_AUDIO,
100
+ SemanticConvetion.GEN_AI_REQUEST_MODEL:
101
+ kwargs.get("model", "eleven_multilingual_v2")
102
+ }
103
+
104
+ metrics["genai_requests"].add(1, attributes)
105
+ metrics["genai_cost"].record(cost, attributes)
106
+
107
+ # Return original response
108
+ return response
109
+
110
+ except Exception as e:
111
+ handle_exception(span, e)
112
+ logger.error("Error in trace creation: %s", e)
113
+
114
+ # Return original response
115
+ return response
116
+
117
+ return wrapper
openlit/instrumentation/elevenlabs/elevenlabs.py ADDED
@@ -0,0 +1,123 @@
1
+ # pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, possibly-used-before-assignment
2
+ """
3
+ Module for monitoring Ollama API calls.
4
+ """
5
+
6
+ import logging
7
+ from opentelemetry.trace import SpanKind, Status, StatusCode
8
+ from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
9
+ from openlit.__helpers import get_audio_model_cost
10
+ from openlit.__helpers import handle_exception
11
+ from openlit.semcov import SemanticConvetion
12
+
13
+ # Initialize logger for logging potential issues and operations
14
+ logger = logging.getLogger(__name__)
15
+
16
+ def generate(gen_ai_endpoint, version, environment, application_name,
17
+ tracer, pricing_info, trace_content, metrics, disable_metrics):
18
+ """
19
+ Generates a telemetry wrapper for creating speech audio to collect metrics.
20
+
21
+ Args:
22
+ gen_ai_endpoint: Endpoint identifier for logging and tracing.
23
+ version: Version of the monitoring package.
24
+ environment: Deployment environment (e.g., production, staging).
25
+ application_name: Name of the application using the ElevenLabs API.
26
+ tracer: OpenTelemetry tracer for creating spans.
27
+ pricing_info: Information used for calculating the cost of generating speech audio.
28
+ trace_content: Flag indicating whether to trace the input text and generated audio.
29
+
30
+ Returns:
31
+ A function that wraps the speech audio creation method to add telemetry.
32
+ """
33
+
34
+ def wrapper(wrapped, instance, args, kwargs):
35
+ """
36
+ Wraps the 'generate' API call to add telemetry.
37
+
38
+ This collects metrics such as execution time, cost, and handles errors
39
+ gracefully, adding details to the trace for observability.
40
+
41
+ Args:
42
+ wrapped: The original 'generate' method to be wrapped.
43
+ instance: The instance of the class where the original method is defined.
44
+ args: Positional arguments for the 'generate' method.
45
+ kwargs: Keyword arguments for the 'generate' method.
46
+
47
+ Returns:
48
+ The response from the original 'generate' method.
49
+ """
50
+
51
+ with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
52
+ response = wrapped(*args, **kwargs)
53
+
54
+ try:
55
+ # Calculate cost of the operation
56
+ cost = get_audio_model_cost(kwargs.get("model", "eleven_multilingual_v2"),
57
+ pricing_info, kwargs.get("text", ""))
58
+
59
+ # Set Span attributes
60
+ span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
61
+ span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
62
+ SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS)
63
+ span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
64
+ SemanticConvetion.GEN_AI_TYPE_AUDIO)
65
+ span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
66
+ gen_ai_endpoint)
67
+ span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
68
+ environment)
69
+ span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
70
+ application_name)
71
+ if gen_ai_endpoint == "elevenlabs.generate":
72
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
73
+ kwargs.get("model", "eleven_multilingual_v2"))
74
+ if isinstance(kwargs.get("voice", "Rachel"), str):
75
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_VOICE,
76
+ kwargs.get("voice", "Rachel"))
77
+ else:
78
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
79
+ kwargs.get("model_id", "eleven_multilingual_v2"))
80
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_VOICE,
81
+ kwargs.get("voice_id", ""))
82
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT,
83
+ kwargs.get("output_format", "mp3"))
84
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_SETTINGS,
85
+ str(kwargs.get("voice_settings", "")))
86
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
87
+ cost)
88
+ if trace_content:
89
+ span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
90
+ str(kwargs.get("text", "")))
91
+
92
+ span.set_status(Status(StatusCode.OK))
93
+
94
+ if disable_metrics is False:
95
+ attributes = {
96
+ TELEMETRY_SDK_NAME:
97
+ "openlit",
98
+ SemanticConvetion.GEN_AI_APPLICATION_NAME:
99
+ application_name,
100
+ SemanticConvetion.GEN_AI_SYSTEM:
101
+ SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS,
102
+ SemanticConvetion.GEN_AI_ENVIRONMENT:
103
+ environment,
104
+ SemanticConvetion.GEN_AI_TYPE:
105
+ SemanticConvetion.GEN_AI_TYPE_AUDIO,
106
+ SemanticConvetion.GEN_AI_REQUEST_MODEL:
107
+ kwargs.get("model", "eleven_multilingual_v2")
108
+ }
109
+
110
+ metrics["genai_requests"].add(1, attributes)
111
+ metrics["genai_cost"].record(cost, attributes)
112
+
113
+ # Return original response
114
+ return response
115
+
116
+ except Exception as e:
117
+ handle_exception(span, e)
118
+ logger.error("Error in trace creation: %s", e)
119
+
120
+ # Return original response
121
+ return response
122
+
123
+ return wrapper
openlit/instrumentation/groq/async_groq.py CHANGED
@@ -119,15 +119,15 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
119
119
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
120
120
  kwargs.get("user", ""))
121
121
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
122
- kwargs.get("top_p", 1))
122
+ kwargs.get("top_p", 1.0))
123
123
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
124
- kwargs.get("max_tokens", ""))
124
+ kwargs.get("max_tokens", -1))
125
125
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
126
- kwargs.get("temperature", 1))
126
+ kwargs.get("temperature", 1.0))
127
127
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
128
- kwargs.get("presence_penalty", 0))
128
+ kwargs.get("presence_penalty", 0.0))
129
129
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
130
- kwargs.get("frequency_penalty", 0))
130
+ kwargs.get("frequency_penalty", 0.0))
131
131
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
132
132
  kwargs.get("seed", ""))
133
133
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -221,17 +221,17 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
221
221
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
222
222
  kwargs.get("model", "llama3-8b-8192"))
223
223
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
224
- kwargs.get("top_p", 1))
224
+ kwargs.get("top_p", 1.0))
225
225
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
226
- kwargs.get("max_tokens", ""))
226
+ kwargs.get("max_tokens", -1))
227
227
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
228
228
  kwargs.get("name", ""))
229
229
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
230
- kwargs.get("temperature", 1))
230
+ kwargs.get("temperature", 1.0))
231
231
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
232
- kwargs.get("presence_penalty", 0))
232
+ kwargs.get("presence_penalty", 0.0))
233
233
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
234
- kwargs.get("frequency_penalty", 0))
234
+ kwargs.get("frequency_penalty", 0.0))
235
235
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
236
236
  kwargs.get("seed", ""))
237
237
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/groq/groq.py CHANGED
@@ -119,15 +119,15 @@ def chat(gen_ai_endpoint, version, environment, application_name,
119
119
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
120
120
  kwargs.get("user", ""))
121
121
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
122
- kwargs.get("top_p", 1))
122
+ kwargs.get("top_p", 1.0))
123
123
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
124
- kwargs.get("max_tokens", ""))
124
+ kwargs.get("max_tokens", -1))
125
125
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
126
- kwargs.get("temperature", 1))
126
+ kwargs.get("temperature", 1.0))
127
127
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
128
- kwargs.get("presence_penalty", 0))
128
+ kwargs.get("presence_penalty", 0.0))
129
129
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
130
- kwargs.get("frequency_penalty", 0))
130
+ kwargs.get("frequency_penalty", 0.0))
131
131
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
132
132
  kwargs.get("seed", ""))
133
133
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -221,17 +221,17 @@ def chat(gen_ai_endpoint, version, environment, application_name,
221
221
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
222
222
  kwargs.get("model", "llama3-8b-8192"))
223
223
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
224
- kwargs.get("top_p", 1))
224
+ kwargs.get("top_p", 1.0))
225
225
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
226
- kwargs.get("max_tokens", ""))
226
+ kwargs.get("max_tokens", -1))
227
227
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
228
228
  kwargs.get("name", ""))
229
229
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
230
- kwargs.get("temperature", 1))
230
+ kwargs.get("temperature", 1.0))
231
231
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
232
- kwargs.get("presence_penalty", 0))
232
+ kwargs.get("presence_penalty", 0.0))
233
233
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
234
- kwargs.get("frequency_penalty", 0))
234
+ kwargs.get("frequency_penalty", 0.0))
235
235
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
236
236
  kwargs.get("seed", ""))
237
237
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/mistral/async_mistral.py CHANGED
@@ -95,9 +95,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
95
95
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
96
96
  kwargs.get("temperature", 0.7))
97
97
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
98
- kwargs.get("top_p", 1))
98
+ kwargs.get("top_p", 1.0))
99
99
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
100
- kwargs.get("max_tokens", ""))
100
+ kwargs.get("max_tokens", -1))
101
101
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
102
102
  kwargs.get("random_seed", ""))
103
103
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -252,9 +252,9 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
252
252
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
253
253
  kwargs.get("temperature", 0.7))
254
254
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
255
- kwargs.get("top_p", 1))
255
+ kwargs.get("top_p", 1.0))
256
256
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
257
- kwargs.get("max_tokens", ""))
257
+ kwargs.get("max_tokens", -1))
258
258
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
259
259
  kwargs.get("random_seed", ""))
260
260
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/mistral/mistral.py CHANGED
@@ -94,9 +94,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
94
94
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
95
95
  kwargs.get("temperature", 0.7))
96
96
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
97
- kwargs.get("top_p", 1))
97
+ kwargs.get("top_p", 1.0))
98
98
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
99
- kwargs.get("max_tokens", ""))
99
+ kwargs.get("max_tokens", -1))
100
100
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
101
101
  kwargs.get("random_seed", ""))
102
102
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -251,9 +251,9 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
251
251
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
252
252
  kwargs.get("temperature", 0.7))
253
253
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
254
- kwargs.get("top_p", 1))
254
+ kwargs.get("top_p", 1.0))
255
255
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
256
- kwargs.get("max_tokens", ""))
256
+ kwargs.get("max_tokens", -1))
257
257
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
258
258
  kwargs.get("random_seed", ""))
259
259
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/openai/async_azure_openai.py CHANGED
@@ -125,11 +125,11 @@ def azure_async_chat_completions(gen_ai_endpoint, version, environment, applicat
125
125
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
126
126
  kwargs.get("tool_choice", ""))
127
127
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
128
- kwargs.get("temperature", 1))
128
+ kwargs.get("temperature", 1.0))
129
129
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
130
- kwargs.get("presence_penalty", 0))
130
+ kwargs.get("presence_penalty", 0.0))
131
131
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
132
- kwargs.get("frequency_penalty", 0))
132
+ kwargs.get("frequency_penalty", 0.0))
133
133
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
134
134
  kwargs.get("seed", ""))
135
135
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -228,11 +228,11 @@ def azure_async_chat_completions(gen_ai_endpoint, version, environment, applicat
228
228
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
229
229
  kwargs.get("tool_choice", ""))
230
230
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
231
- kwargs.get("temperature", 1))
231
+ kwargs.get("temperature", 1.0))
232
232
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
233
- kwargs.get("presence_penalty", 0))
233
+ kwargs.get("presence_penalty", 0.0))
234
234
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
235
- kwargs.get("frequency_penalty", 0))
235
+ kwargs.get("frequency_penalty", 0.0))
236
236
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
237
237
  kwargs.get("seed", ""))
238
238
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -425,11 +425,11 @@ def azure_async_completions(gen_ai_endpoint, version, environment, application_n
425
425
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
426
426
  kwargs.get("tool_choice", ""))
427
427
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
428
- kwargs.get("temperature", 1))
428
+ kwargs.get("temperature", 1.0))
429
429
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
430
- kwargs.get("presence_penalty", 0))
430
+ kwargs.get("presence_penalty", 0.0))
431
431
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
432
- kwargs.get("frequency_penalty", 0))
432
+ kwargs.get("frequency_penalty", 0.0))
433
433
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
434
434
  kwargs.get("seed", ""))
435
435
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -509,11 +509,11 @@ def azure_async_completions(gen_ai_endpoint, version, environment, application_n
509
509
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
510
510
  kwargs.get("tool_choice", ""))
511
511
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
512
- kwargs.get("temperature", 1))
512
+ kwargs.get("temperature", 1.0))
513
513
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
514
- kwargs.get("presence_penalty", 0))
514
+ kwargs.get("presence_penalty", 0.0))
515
515
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
516
- kwargs.get("frequency_penalty", 0))
516
+ kwargs.get("frequency_penalty", 0.0))
517
517
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
518
518
  kwargs.get("seed", ""))
519
519
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/openai/async_openai.py CHANGED
@@ -122,15 +122,15 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
122
122
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
123
123
  kwargs.get("user", ""))
124
124
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
125
- kwargs.get("top_p", 1))
125
+ kwargs.get("top_p", 1.0))
126
126
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
127
- kwargs.get("max_tokens", ""))
127
+ kwargs.get("max_tokens", -1))
128
128
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
129
- kwargs.get("temperature", 1))
129
+ kwargs.get("temperature", 1.0))
130
130
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
131
- kwargs.get("presence_penalty", 0))
131
+ kwargs.get("presence_penalty", 0.0))
132
132
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
133
- kwargs.get("frequency_penalty", 0))
133
+ kwargs.get("frequency_penalty", 0.0))
134
134
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
135
135
  kwargs.get("seed", ""))
136
136
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -224,17 +224,17 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
224
224
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
225
225
  kwargs.get("model", "gpt-3.5-turbo"))
226
226
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
227
- kwargs.get("top_p", 1))
227
+ kwargs.get("top_p", 1.0))
228
228
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
229
- kwargs.get("max_tokens", ""))
229
+ kwargs.get("max_tokens", -1))
230
230
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
231
231
  kwargs.get("user", ""))
232
232
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
233
- kwargs.get("temperature", 1))
233
+ kwargs.get("temperature", 1.0))
234
234
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
235
- kwargs.get("presence_penalty", 0))
235
+ kwargs.get("presence_penalty", 0.0))
236
236
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
237
- kwargs.get("frequency_penalty", 0))
237
+ kwargs.get("frequency_penalty", 0.0))
238
238
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
239
239
  kwargs.get("seed", ""))
240
240
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/openai/azure_openai.py CHANGED
@@ -125,11 +125,11 @@ def azure_chat_completions(gen_ai_endpoint, version, environment, application_na
125
125
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
126
126
  kwargs.get("tool_choice", ""))
127
127
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
128
- kwargs.get("temperature", 1))
128
+ kwargs.get("temperature", 1.0))
129
129
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
130
- kwargs.get("presence_penalty", 0))
130
+ kwargs.get("presence_penalty", 0.0))
131
131
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
132
- kwargs.get("frequency_penalty", 0))
132
+ kwargs.get("frequency_penalty", 0.0))
133
133
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
134
134
  kwargs.get("seed", ""))
135
135
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -228,11 +228,11 @@ def azure_chat_completions(gen_ai_endpoint, version, environment, application_na
228
228
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
229
229
  kwargs.get("tool_choice", ""))
230
230
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
231
- kwargs.get("temperature", 1))
231
+ kwargs.get("temperature", 1.0))
232
232
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
233
- kwargs.get("presence_penalty", 0))
233
+ kwargs.get("presence_penalty", 0.0))
234
234
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
235
- kwargs.get("frequency_penalty", 0))
235
+ kwargs.get("frequency_penalty", 0.0))
236
236
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
237
237
  kwargs.get("seed", ""))
238
238
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -423,11 +423,11 @@ def azure_completions(gen_ai_endpoint, version, environment, application_name,
423
423
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
424
424
  kwargs.get("tool_choice", ""))
425
425
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
426
- kwargs.get("temperature", 1))
426
+ kwargs.get("temperature", 1.0))
427
427
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
428
- kwargs.get("presence_penalty", 0))
428
+ kwargs.get("presence_penalty", 0.0))
429
429
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
430
- kwargs.get("frequency_penalty", 0))
430
+ kwargs.get("frequency_penalty", 0.0))
431
431
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
432
432
  kwargs.get("seed", ""))
433
433
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -507,11 +507,11 @@ def azure_completions(gen_ai_endpoint, version, environment, application_name,
507
507
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
508
508
  kwargs.get("tool_choice", ""))
509
509
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
510
- kwargs.get("temperature", 1))
510
+ kwargs.get("temperature", 1.0))
511
511
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
512
- kwargs.get("presence_penalty", 0))
512
+ kwargs.get("presence_penalty", 0.0))
513
513
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
514
- kwargs.get("frequency_penalty", 0))
514
+ kwargs.get("frequency_penalty", 0.0))
515
515
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
516
516
  kwargs.get("seed", ""))
517
517
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/openai/openai.py CHANGED
@@ -123,15 +123,15 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
123
123
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
124
124
  kwargs.get("user", ""))
125
125
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
126
- kwargs.get("top_p", 1))
126
+ kwargs.get("top_p", 1.0))
127
127
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
128
- kwargs.get("max_tokens", ""))
128
+ kwargs.get("max_tokens", -1))
129
129
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
130
- kwargs.get("temperature", 1))
130
+ kwargs.get("temperature", 1.0))
131
131
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
132
- kwargs.get("presence_penalty", 0))
132
+ kwargs.get("presence_penalty", 0.0))
133
133
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
134
- kwargs.get("frequency_penalty", 0))
134
+ kwargs.get("frequency_penalty", 0.0))
135
135
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
136
136
  kwargs.get("seed", ""))
137
137
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
@@ -225,17 +225,17 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
225
225
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
226
226
  kwargs.get("model", "gpt-3.5-turbo"))
227
227
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
228
- kwargs.get("top_p", 1))
228
+ kwargs.get("top_p", 1.0))
229
229
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
230
- kwargs.get("max_tokens", ""))
230
+ kwargs.get("max_tokens", -1))
231
231
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
232
232
  kwargs.get("user", ""))
233
233
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
234
- kwargs.get("temperature", 1))
234
+ kwargs.get("temperature", 1.0))
235
235
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
236
- kwargs.get("presence_penalty", 0))
236
+ kwargs.get("presence_penalty", 0.0))
237
237
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
238
- kwargs.get("frequency_penalty", 0))
238
+ kwargs.get("frequency_penalty", 0.0))
239
239
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
240
240
  kwargs.get("seed", ""))
241
241
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
openlit/instrumentation/transformers/transformers.py CHANGED
@@ -87,7 +87,7 @@ def text_wrap(gen_ai_endpoint, version, environment, application_name,
87
87
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
88
88
  forward_params.get("top_p", "null"))
89
89
  span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
90
- forward_params.get("max_length", "null"))
90
+ forward_params.get("max_length", -1))
91
91
  span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
92
92
  prompt)
93
93
  if trace_content: