Add OpenAI embeddings instrumentation

[drewby]

Description

This PR adds instrumentation for OpenAI's embeddings API in the GenAI instrumentation suite. The implementation follows the OpenTelemetry semantic conventions for generative AI systems and provides automatic instrumentation for the OpenAI Python client when using embeddings functionality.

The implementation captures important metadata about embedding operations including model, dimensions, and relevant timing information while respecting sensitive data handling practices.

• Added instrumentation for both synchronous and asynchronous OpenAI embedding API calls
• Implemented span and metrics using existing attributes, with two new custom:
  • ai.embedding.dimensions - Number of dimensions in the embedding vectors
  • ai.embedding.encoding_format - The encoding format of the embedding vectors response (base64 or float)
• Capturing input text content (disabled by default for privacy)
• Added a usage example called embeddings

Type of change

• New feature (non-breaking change which adds functionality)

How Has This Been Tested?

• Unit tests using mock responses to verify proper span creation and attribute population
• Integration tests with the OpenAI client against a mock server
• Manual testing using examples/embeddings with real OpenAI service

Does This PR Require a Core Repo Change?

• Yes. - Link to PR:
• No.

Checklist:

See contributing.md for styleguide, changelog guidelines, and more.

• Followed the style guidelines of this project
• Changelogs have been updated
• Unit tests have been added
• Documentation has been updated

[lmolkova]

thank you!!!

[drewby]

@lmolkova @xrmx, I think I've removed all of the Event related code, so now this is just recording the Span and Metrics.

[anuraaga]

Hey - just wondering what's the status of this PR as interested in having embeddings instrumented here. If there's anything I can do to help on it, let me know.

[drewby]

@pmcollins @xrmx @lmolkova Thank You! I'm working on it and wil have updates soon.

[drewby]

@xrmx @lmolkova I've made several updates based on the feedback. I hope I addressed most concerns. There is one check not passing but seems unrelated to openai library.

[aabmass]

@xrmx can you take another pass?

[Copilot]

Pull Request Overview

This PR adds comprehensive instrumentation for OpenAI's embeddings API to the existing OpenTelemetry instrumentation suite. The implementation provides automatic tracing and metrics collection for both synchronous and asynchronous embedding operations following OpenTelemetry semantic conventions.

• Adds support for tracing OpenAI embeddings API calls with proper span attributes and metrics
• Introduces new custom attributes for embedding-specific metadata (dimensions and encoding format)
• Includes comprehensive test coverage with both unit and integration tests

Reviewed Changes

Copilot reviewed 26 out of 28 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
src/opentelemetry/instrumentation/openai_v2/__init__.py	Adds instrumentation hooks for embeddings endpoints
src/opentelemetry/instrumentation/openai_v2/patch.py	Implements tracing logic for sync/async embeddings operations
src/opentelemetry/instrumentation/openai_v2/utils.py	Adds embeddings-specific request attribute handling
tests/test_embeddings.py	Comprehensive test suite for synchronous embeddings API
tests/test_async_embeddings.py	Test coverage for asynchronous embeddings functionality
tests/test_utils.py	Shared test utilities for consistent assertion patterns

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[xrmx]

Suggested change

	opentelemetry-instrumentation-openai-v2~=2.1b0
	opentelemetry-instrumentation-openai-v2~=2.2b0

[xrmx]

A couple of comments but LGTM.

[xrmx]

We need to guard against instances of openai.NotGiven (available since 1.13.4) and so add one test for sync and one for async checking that for the encoding_format parameter . We have another PR adding guards for chat completions #3760

[xrmx]

I guess this should be GenAIAttributes.GEN_AI_OPENAI_REQUEST_SERVICE_TIER, care to fix while you are at it? Don't want to create conflicts on another PR :) With a separate entry in the changelog though. If you don't want to do that in this PR we can do it later, no worries.

[xrmx]

Also I would ignore copilot comments.