kitio.aiAdded Oct 28, 2026
Lakera
FreemiumSecurity platform protecting GenAI apps and agents from prompt attacks
What is Lakera?
Lakera protects generative AI applications and agents against prompt injection, data leaks, toxic output and other AI-specific attacks, with runtime checks that add under 50 ms of latency. It also helps companies govern employee AI use and offers AI red teaming.
Lakera's runtime protection sits between users, models and tools to screen inputs and outputs for direct and indirect prompt injection, sensitive data and policy violations in more than 100 languages. Use cases include conversational agents, document and RAG agents, GenAI gateways and agents connected through MCP.
Its threat intelligence draws on Gandalf, a public prompt-injection game played by more than a million people. A risk-based red teaming product and an expert-led red teaming service test applications before launch, and developers can start on a free tier.
Features
- Prompt injection defence
- Data leak prevention
- Agent and MCP security
- Employee AI governance
- AI red teaming
- Low-latency runtime API
Tags
Releases
+ Submit new release- Listed
Lakera was added to the directory.
Organization
Lakera
🇨🇭 Switzerland
Listed by kitio.ai · unverified owner
Pricing
- Pricing model
- Freemium
- Paid options from
- Not listed
Sponsored
Your AI tool here
A Featured launch puts your tool in this spot and on the home page for a week.
Top Lakera alternatives
View allModels & APIs
Automated red teaming and security testing for AI models and agents
If you liked Lakera
Open-source CLI for testing prompts and red-teaming AI apps
FreemiumModels & APIsMade in United StatesSecurity and governance platform for enterprise AI models and agents
PaidModels & APIsMade in United States- JaxonNew
Rule-checking layer that returns a TRUE, FALSE or UNKNOWN verdict on AI output
PaidModels & APIs
Pros and cons
Pros
- Low-latency runtime checks
- Covers 100+ languages
- Free tier for developers
Cons
- Enterprise pricing isn't public
- Adds another service to your AI stack
- Some products are sales-led
Prompts & results
No prompts yet. Add your own prompt and output to help others use this tool.
Share a prompt
Reviews
Q&A
What is Lakera?
Lakera protects generative AI applications and agents against prompt injection, data leaks, toxic output and other AI-specific attacks, with runtime checks that add under 50 ms of latency. It also helps companies govern employee AI use and offers AI red teaming.
What attacks does Lakera protect against?
Prompt injection (direct and indirect), AI data leaks, toxic content, compliance risks and multilingual or multimodal attacks.
How much does Lakera cost?
Lakera is freemium.
Is there a free version of Lakera?
Yes. Lakera has a free plan, and paid plans add more usage and features.
What are the main features of Lakera?
Prompt injection defence, Data leak prevention, Agent and MCP security, Employee AI governance, AI red teaming, Low-latency runtime API.
Who makes Lakera?
Lakera is made by Lakera.
What are the best alternatives to Lakera?
Popular alternatives include Dynamo AI, Mindgard, Giskard.