Lakera
Runtime security for LLM and agent applications
Lakera is profiled here as a Guardrails tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
builds runtime security for GenAI applications and was founded in Zurich in 2021 by David Haber, Mateo Rojas-Carulla, and Matthias Kraft. Check Point Software acquired the company in 2025, making Lakera the anchor of its end-to-end AI security stack, with the Zurich office serving as a global AI security research center. Lakera Guard screens prompts aLakera nd model outputs in real time before they reach users or tools. The company built Gandalf, a prompt injection game played by millions, and the resulting attack data trains the detection models behind Lakera Guard.
Key Capabilities:
Prompt injection and jailbreak detection at low latency
Data leakage and PII screening for model inputs and outputs
Content moderation policies for harmful output
Lakera Red for continuous adversarial red teaming
Threat intelligence drawn from the Gandalf adversarial network
REST API with SaaS and on-premises deployment
Alternative tools
- LlamaFirewall
Open-source guardrail framework for securing AI agents
- HiddenLayer
Security platform for protecting machine learning models
- Arthur
Real-time guardrails and monitoring for AI applications
- Guardrails AI
Open-source validation framework for LLM inputs and outputs
- NeMo Guardrails
Enforce safety policies across live LLM conversations using a programmable rail architecture