Garak
garak is an open-source LLM vulnerability scanner, with dozens of plugins and thousands of prompts that test large language model security.
garak is an open-source LLM vulnerability scanner, with dozens of plugins and thousands of prompts that test large language model security.
CalypsoAI
Define and deploy agile data security, threat management, and governance for AI models, apps, and agents.
Define and deploy agile data security, threat management, and governance for AI models, apps, and agents. Safeguard AI systems from evolving threats like prompt injection and jailbreaks. Choose from preset guardrails or create bespoke policies for specific use cases. Detect and prevent data leakage, compliance failures, and policy violations at runtime. Ensure regulatory compliance, obstruct harmful outputs, and enforce restrictions on model and agent privileges. Achieve continuous visibility and traceability across all AI interactions. AI expands the attack surface in every direction. To maintain your security posture, teams need solutions that balance efficient workflow automation with strategic prioritization and continuous protection against evolving threats. F5 AI Guardrails meets the evolving needs of AI security by providing scalable data governance, augmented threat management, and risk auditing for present and future challenges. Apply tailored risk evaluation frameworks to public foundational models and in-house models alike. Inspect AI interactions across models and apps, and provide real-time protection for DLP and policy violations. Ensure enterprise-wide policy alignment with automated auditing templates for GDPR, HIPAA, EUAIA, and more. Rapidly translate insights from F5 AI Red Team and agentic threat intelligence into active defense strategy. Dynamic model routing to avoid failover states, and maintain performance without compromising security. Avoid detrimental outputs with content moderation filters for toxic, biased, or inaccurate content. Safeguard frontier models with preset configurations for the most popular enterprise and open-source AI. See how F5 AI Guardrails performed against 17,733 adversarial test cases, independently validated by SecureIQLab.
Garak
CalypsoAI
Garak
CalypsoAI
CalypsoAI (6)
Only in CalypsoAI (10)
Garak
No screenshots
CalypsoAI
Garak
CalypsoAI
