Conversation
Surface a sliding-window tok/s reading on the transcript stream-health strip so users can tell whether the model stream is healthy or stalled. Prefer provider output usage when available; otherwise estimate from visible text and label as approximate. Fixes vastsa#394.
Keep the chip hidden until positive output (no false TTFT 0), reset the sample window on estimate→provider handoff when provider is lower, sample on a stable 250ms interval via input refs, and treat the ticking rate as decorative for a11y. Mirror D428 in zh-CN specs and use FR jetons/s.
Resolve conflicts with main: - ChatTranscript/ActivityGroup: re-apply the live tok/s chip on top of the always-mounted runtime status lane and the extracted ActivityItems; drop the separate streaming-only indicator since the lane now stays mounted for the whole running turn. - Spec/E2E/decision-log: keep upstream entries; renumber this change's decision from D428 to D637 (D428 is now taken upstream). - i18n: add the live throughput keys to the new pt-BR locale.
Resolve decision-log conflicts: keep upstream entries; renumber this change's decision from D637 to D639 (D637 and D638 are now taken upstream).
Contributor
Author
|
Merged the latest What conflicted and how it was resolved
Checks run locally (Node 22.23, pnpm 12.8.1)
|
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #394
Summary
tok/s) on the transcript stream-health strip while a turn is running (working / run-activity indicators, plus a compact chip while answer tokens stream).outputTokensmid-stream; otherwise estimate from visible thinking+answer text (four code points ≈ one token) and label as approximate (≈).Algorithm
useLiveTokenRatekeeps content/thinking/outputTokensin refs so stream deltas do not tear down the timer).undefined(chip hidden)—never a false0 tok/s.0so reconnect/stall is visible.0without≈.Review fixes (this push)
aria-hidden/ decorative (mirrorsworking-elapsed); streaming-only chip is not a polite live region.tokensPerSecond !== undefined(not always-truthy{tokenRate ? …}).jetons/s.Tests
pnpm typecheck/pnpm lintinapps/desktop— greennode --test test/streaming-token-rate.test.mjs test/active-turn-surface.test.mjs— green (TTFT undefined, estimate→provider reset, hook interval deps, a11y/guard contracts)