garak
Comparisons anchored on garak, NVIDIA's Apache-2.0 LLM vulnerability scanner, against other open-source AI red teaming tools.
AI security tooling
Comparisons of open-source tools for testing LLM applications and adding runtime guardrails, judged on license, maintainer, attack coverage, how they fit into CI and what it takes to run them on your own systems.
promptfoo
Comparisons anchored on promptfoo, the MIT-licensed LLM eval and red teaming CLI now part of OpenAI.
NeMo Guardrails
Comparisons anchored on NeMo Guardrails, NVIDIA's Apache-2.0 runtime guardrail toolkit. LLM Guard is archived and not compared.
LLM Guard
Comparisons that start from LLM Guard, the open-source toolkit of input and output scanners for LLM applications, against other runtime guardrail libraries.
AI security tooling compared
Roundups of open-source tools for testing LLM applications and guarding them at runtime, judged on license, maintainer, coverage and fit with CI and production systems.
PyRIT
Comparisons anchored on PyRIT, Microsoft's MIT-licensed framework for scripting multi-turn attacks on generative AI systems, including the Azure service built on it.
Guardrails AI
Comparisons anchored on Guardrails AI, the Apache-2.0 library of input and output validators and structured output, and how it is deployed.
Meta Purple Llama
Comparisons inside Meta's Purple Llama safeguards: Llama Guard, Prompt Guard and LlamaFirewall, and how their versions and licenses differ.
Adversarial Robustness Toolbox (ART)
Comparisons anchored on ART, the LF AI & Data library for testing classic machine learning models against evasion, poisoning, extraction and inference attacks.
ModelScan
Comparisons anchored on ModelScan, Protect AI's Apache-2.0 scanner for unsafe code in serialized model files, against other open-source pickle scanners.
