ta
TanStack
OSS
ta
Pipelines
Nx runs
Insights
Compare tasks
Analytics
Sign in
Toggle Sidebar
ai
ai
Pipelines
fix(ai-compaction): key native checkpoints by model, keep native on after transient failures - The native checkpoint key now always holds the model, also with an explicit `strategyKey`. Before, model B could reuse model A's checkpoint, and the Responses adapter dropped the foreign compaction item, so the compacted context was silently lost. - Only a 404, 405, or 501 (no compaction endpoint) turns native compaction off for the adapter and model. Before, any failure did, so one 429, one network error, or one thread's 400 turned it off for every thread until the process restarted.
nx run-many --targets=build --exclude=examples/**,testing/**
⌘K
fix(ai-compaction): key native checkpoints by model, keep native on after transient failures
Succeeded
in
2m 23s
Started
Linux
4 CPU cores
Copy command
nx run-many --targets=build --exclude=examples/**,testing/**
1729
65ef720
GitHub
Open on GitHub
Tasks
Resource Usage
ta
TanStack
OSS