Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Azure AI Inference SDK - Beta 2 updates #36163

Merged
merged 10 commits into from
Jun 22, 2024
Merged

Conversation

dargilco
Copy link
Member

@dargilco dargilco commented Jun 19, 2024

The main reason for this release, shortly after the first release:

  • Add strongly-typed model as an optional input argument to the complete method of ChatCompletionsClient. This is required for a high-visiblity project. For this project, developers must set model.

Breaking change (noted in CHANGELOG.md):

  • The field input_tokens was removed from class EmbeddingsUsage, as this was never defined in the
    REST API and the service never returned this value.

Other changes in this release:

  • Addressing some test dept (work in progress)
    • Add tests for setting model_extras for sync and async clients. Make sure the additional parameters appear at the root of the JSON request payload, and make sure the unknown_parameters HTTP request header was set to pass_through.
    • Add tests to validate serialization of a dummy chat completion request that includes all type of input objects. This is a regression test (no service response needed), as the test looks at the JSON request payload and compared to a hard-coded expected string, that was previously verified by hand. This test includes the new model argument, as well as all other arguments defined by the REST API. It will catch any regressions in hand-written code.
  • Update ref docs to remove mentioning of the old extras input argument to chat completions in hand-written code. The name was changed to model_extras before the first release, but looks like we still had some left-over ref-doc comments that describe the no-longer-existing argument.
  • Remove unused function from the sample sample_chat_completions_with_image_data.py. Forgot to do that in the first release.
  • Minor changes to root README.md
  • Indicate that complete method with stream=True returns Iterable[StreamingChatCompletionsUpdate] for
    the synchronous ChatComletionsClient, and Iterable[StreamingChatCompletionsUpdate] for the asynchronous
    ChatCompletionsClient. Per feedback from Anna T.
  • Update environment variable names used by sample code and test to start with "AZURE_AI" as common elsewhere, per feedback from Rob C.

@github-actions github-actions bot added the AI Model Inference Issues related to the client library for Azure AI Model Inference (\sdk\ai\azure-ai-inference) label Jun 19, 2024
@dargilco dargilco self-assigned this Jun 19, 2024
@azure-sdk
Copy link
Collaborator

API change check

API changes are not detected in this pull request.

@dargilco dargilco marked this pull request as ready for review June 21, 2024 14:15
@dargilco dargilco requested review from annatisch and johanste June 21, 2024 14:16
@dargilco dargilco requested a review from robch June 21, 2024 14:32
@dargilco dargilco enabled auto-merge (squash) June 21, 2024 17:54
@dargilco dargilco merged commit 444ed8b into main Jun 22, 2024
17 checks passed
@dargilco dargilco deleted the dargilco/azure-ai-inference-6-14 branch June 22, 2024 00:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
AI Model Inference Issues related to the client library for Azure AI Model Inference (\sdk\ai\azure-ai-inference)
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants