ta
TanStack
OSS
ta
Pipelines
Nx runs
Insights
Compare tasks
Analytics
Sign in
Toggle Sidebar
ai
ai
Pipelines
fix(ai-client): coalesce the per-chunk snapshot storm while streaming A large tool-call stream re-rendered every useChat hook once per chunk. A Code Mode program arrives as thousands of TOOL_CALL_ARGS deltas in one burst, so the client published each delta to the snapshot atom synchronously and the tab locked up (the SSE read runs on the same thread and could abort). Message publishes made while the client processes a chunk now coalesce to about one per animation frame (setTimeout fallback off-DOM). Discrete changes (setMessages, clear, reload, status) stay synchronous, and getSnapshot() flushes any pending streamed messages so it is never stale. In development, the devtools bridge's per-event snapshot emit is likewise coalesced to one per microtask. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
nx run-many --targets=build --exclude=examples/**,testing/**
⌘K
fix(ai-client): coalesce the per-chunk snapshot storm while streaming
Succeeded
in
4s
Started
Linux
4 CPU cores
read-write
access token used
Copy command
nx run-many --targets=build --exclude=examples/**,testing/**
1956
ebc12c9
GitHub
Open on GitHub
Tasks
Resource Usage
ta
TanStack
OSS