MLE-5325: docs(audio): drop task field from verbose_json schemas#271
Merged
rishabh-bhargava merged 1 commit intomainfrom May 7, 2026
Merged
MLE-5325: docs(audio): drop task field from verbose_json schemas#271rishabh-bhargava merged 1 commit intomainfrom
rishabh-bhargava merged 1 commit intomainfrom
Conversation
The task field was specified as required in both AudioTranscriptionVerboseJsonResponse and AudioTranslationVerboseJsonResponse but was never built by inference-pop or tvllm. The endpoint URL already discriminates the task (/audio/transcriptions vs /audio/translations), so the field carries no information the caller doesn't already have. Dropping from the spec rather than implementing it; this aligns the documented contract with what the API actually returns today. Note: anyone using strict OpenAI-compat SDKs (openai-python TranscriptionVerbose) that expects `task` will need to skip strict validation. This was the existing state already, since the field was never returned.
✱ Stainless preview buildsThis PR will update the go openapi python terraform typescript
|
sbeurnier
approved these changes
May 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Drops the
taskfield from bothAudioTranscriptionVerboseJsonResponseandAudioTranslationVerboseJsonResponse. The field was specified asrequiredin the spec but was never actually built by inference-pop or tvllm — confirmed via code trace (tvllm/inference_worker/transcription.py:1007-1013andinference-pop/src/v1/audio.ts:708-713both populate the response without it).This aligns the documented contract with the API's actual behavior today. Closes MLE-5325.
Why drop instead of implement?
The endpoint URL already discriminates the task (
/audio/transcriptionsvs/audio/translations), so thetaskresponse field carries no information the caller doesn't already have. OpenAI returns it because their single Whisper endpoint can do either task depending on internal flags; Together split the URLs, so the field is functionally redundant.Compatibility note
Anyone using strict OpenAI-compat SDKs (e.g.
openai-python'sTranscriptionVerbosemodel) that expects ataskfield will need to skip strict validation. This is the existing state — the field was never returned — so this is just documentation catching up to reality.Test plan
mintlify-docs/openapi.yaml).taskin the verbose_json context (will verify before this merges).🤖 Generated with Claude Code