All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Fix
StreamWrappermissing.headersand other attributes when usingwith_raw_responsestreaming (#4113) - Add opt-in support for latest experimental semantic conventions (v1.37.0). Set
OTEL_SEMCONV_STABILITY_OPT_INtogen_ai_latest_experimentalto enable. Add dependency onopentelemetry-util-genaipypi package. (#3715) - Add wrappers for OpenAI Responses API streams and response stream managers (#4280)
- Add async wrappers for OpenAI Responses API streams and response stream managers (#4325)
- Fix
AttributeErrorwhen handlingLegacyAPIResponse(fromwith_raw_response) (#4017) - Add support for chat completions choice count and stop sequences span attributes (#4028)
- Fix crash with streaming
with_raw_response(#4033) - Bump to 1.30.0 semconv schema:
gen_ai.request.seedinstead ofgen_ai.openai.request.seed(#4036)
- Fix service tier attribute names: use
GEN_AI_OPENAI_REQUEST_SERVICE_TIERfor request attributes andGEN_AI_OPENAI_RESPONSE_SERVICE_TIERfor response attributes. (#3920) - Added support for OpenAI embeddings instrumentation (#3461)
- Record prompt and completion events regardless of span sampling decision. (#3226)
- Filter out attributes with the value of NotGiven instances (#3760)
- Migrate off the deprecated events API to use the logs API (#3625)
- Coerce openai response_format to semconv format (#3073)
- Add example to
opentelemetry-instrumentation-openai-v2(#3006) - Support for
AsyncOpenAI/AsyncCompletions(#2984) - Add metrics (#3180)
-
Use generic
OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENTenvironment variable to control if content of prompt, completion, and other messages is captured. (#2947) -
Update OpenAI instrumentation to Semantic Conventions v1.28.0: add new attributes and switch prompts and completions to log-based events. (#2925)
-
Initial OpenAI instrumentation (#2759)