vouch 0.1.0 copy "vouch: ^0.1.0" to clipboard
vouch: ^0.1.0 copied to clipboard

Catch silent on-device LLM regressions: freeze a baseline of your eval suite, diff every model swap against it, and gate CI on what changed.

Changelog #

0.1.0 #

  • Initial release: baseline snapshots of llm_replay_eval reports, regression diffing across model swaps, and the expectNoRegression CI gate.
0
likes
160
points
110
downloads

Documentation

API reference

Publisher

unverified uploader

Weekly Downloads

Catch silent on-device LLM regressions: freeze a baseline of your eval suite, diff every model swap against it, and gate CI on what changed.

Repository (GitHub)
View/report issues

Topics

#llm #testing #regression #ai #on-device

License

MIT (license)

Dependencies

flutter, llm_replay_eval

More

Packages that depend on vouch