falsifyai is an open-source command-line tool for reliability testing of AI systems. It enables researchers to perturb inputs, preserve replayable evidence, and compare model reliability across changes. Designed for AI researchers and engineers focused on model evaluation and safety.
falsifyai is a LLM eval & observability product. It focuses on testing and validating the reliability of AI models by perturbing inputs and comparing model outputs. It is built as an open-source project for AI researchers. falsifyai is open source under the Apache-2.0 license. falsifyai is available on the command line, and it can be self-hosted.
Eric Zhou builds and maintains falsifyai, and the product first shipped in 2026. Development happens publicly on GitHub with 64 commits in the last 90 days. PulseGate's similarity index finds few close equivalents — falsifyai occupies a relatively distinct niche. Key capabilities include input perturbation, replayable evidence, and model diffing.
Latest indexed changes and source events
Other apps tracked under the same category.
