Promptfoo

Latest version: v0.1.0

Safety actively analyzes 706267 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 23 of 28

0.18.1

What's Changed
* Fix issue with writing latest results
* Add support for llama.cpp server by typpo in https://github.com/promptfoo/promptfoo/pull/94


**Full Changelog**: https://github.com/promptfoo/promptfoo/compare/0.18.0...0.18.1

0.18.0

What's Changed
* Promptfoo Theoriesby Skylertodd in https://github.com/promptfoo/promptfoo/pull/89 - allow datasets to be associated with specific set of tests allowing the ability to execute many data driven tests without the requirement of copying a test over and over for each data set.
* Add `python` assertion type that runs python code and returns the output by typpo in https://github.com/promptfoo/promptfoo/pull/78
* Add support for native function ApiProviders and Assertions by typpo in https://github.com/promptfoo/promptfoo/pull/93
* Fix ANSI escape codes in console table
* Fix broken symlink handling for latest output
* Fix eval history cleanup

New Contributors
* Skylertodd made their first contribution in https://github.com/promptfoo/promptfoo/pull/89

**Full Changelog**: https://github.com/promptfoo/promptfoo/compare/0.17.9...0.18.0

0.17.9

What's Changed
* Record latency in results by typpo in https://github.com/promptfoo/promptfoo/pull/85
* Add support for loading test cases from file/directory path by typpo in https://github.com/promptfoo/promptfoo/pull/88
* Fix filepath compatibility issue on Windows


**Full Changelog**: https://github.com/promptfoo/promptfoo/compare/0.17.8...0.17.9

0.17.8

What's Changed
* Additional OpenAI params by typpo in https://github.com/promptfoo/promptfoo/pull/81
* Show recent runs in web viewer by typpo in https://github.com/promptfoo/promptfoo/pull/82
* Add support for postprocessing in test cases by typpo in https://github.com/promptfoo/promptfoo/pull/84
* Fix issue where timeouts for noncritical things (like version check) would crash with error
* Fix bug in evaluator where empty tests were not supported (e.g. when comparing a static prompt on multiple models)


**Full Changelog**: https://github.com/promptfoo/promptfoo/compare/0.17.7...0.17.8

0.17.7

What's Changed
* Add support for provider-specific prompts in the test suite configuration by typpo in https://github.com/promptfoo/promptfoo/pull/76
* Improvements to ReplicateProvider
* Add llama vs gpt bechmark example
* Fixes to llm-rubric prompt grading


**Full Changelog**: https://github.com/promptfoo/promptfoo/compare/0.17.6...0.17.7

0.17.6

What's Changed
* Add "repeat" option to CLI and EvaluateOptions by typpo in https://github.com/promptfoo/promptfoo/pull/71
* Move OpenAI yaml support into shared providers util by typpo in https://github.com/promptfoo/promptfoo/pull/74
* ReplicateProvider implementation by typpo in https://github.com/promptfoo/promptfoo/pull/75
* Web viewer: add support for hiding identical rows
* Darkmode fixes
* Fix colab notebook support


**Full Changelog**: https://github.com/promptfoo/promptfoo/compare/0.17.5...0.17.6

Page 23 of 28

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.