perf(ai): avoid copying Whisper upload bodies#563
Closed
JustYannicc wants to merge 71 commits into
Closed
Conversation
(cherry picked from commit fb7405f)
(cherry picked from commit 290a0f5)
(cherry picked from commit 3739cb0)
(cherry picked from commit c78d0b9)
…ated-integration-stack
…ated-integration-stack # Conflicts: # src/renderer/src/raycast-api/hooks/use-cached-promise.ts
…ated-integration-stack
…ated-integration-stack # Conflicts: # src/renderer/src/raycast-api/list-runtime-renderers.tsx # src/renderer/src/raycast-api/list-runtime.tsx
…ated-integration-stack # Conflicts: # src/renderer/src/CameraExtension.tsx
…ated-integration-stack
…ated-integration-stack # Conflicts: # src/renderer/src/raycast-api/list-runtime.tsx
…ated-integration-stack
…ated-integration-stack
… into codex/perf-consolidated-integration-stack
Collaborator
Author
|
Superseded by consolidated lifecycle/media wave-2 PR: #575 |
Collaborator
Author
|
Closing as superseded by #575. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changed
Why
Large STT uploads previously pushed the original audio buffer into the multipart parts array and then called Buffer.concat(parts), allocating a second full multipart payload before writing to the socket.
Compatibility impact
No user-visible behavior change. Request endpoint, headers, Content-Length, filename/content type selection, optional language field, abort/error handling, and transcript trimming are preserved.
How tested
Performance evidence
scripts/test-ai-provider-whisper-upload.mjs fakes https.request for an 8 MiB audio upload, asserts the exact multipart bytes and Content-Length, and instruments Buffer.concat during the production call. It reports that the legacy concat would allocate an 8,389,124 byte multipart body copy, while the streamed path writes 7 buffers, reuses the 8,388,608 byte audio Buffer, and allocates 516 bytes of framing.
Stack validation
Started from fork/codex/perf-consolidated-integration-stack at 264fb89. Implementation branch: codex/perf-whisper-stream-upload. Commit: 0c60799.
Replaces
None