Open-weight security triage model scoring agent, application and infrastructure events with calibrated probabilities.
Security-One is Superagent's open-weight 27B-parameter model for security decisions, released under the Apache 2.0 license and published on Hugging Face. It is built to run continuously and assess signals across prompts, agent tool calls, code changes, logs and alerts.
Rather than generating free text, the model returns structured probabilities over supplied options, so teams can route suspicious events to stronger models or human reviewers. It is available as open weights and through Superagent's API.
| Attribute | Security-One | Llama Prompt Guard 2 | |
|---|---|---|---|
| Provider |  |  | |
| Founded | 2024 | 2004 | |
| Sells To | Enterprise, Small Business | Consumers, Enterprise | |
| Pricing Model | Recurring, SaaS, Usage-based | Advertising | |
| Ownership | Venture-backed | Public |
Provider
Founded 2024
Sells To Enterprise, Small Business
Pricing Model Recurring, SaaS, Usage-based
Ownership Venture-backed
Provider
Founded 2004
Sells To Consumers, Enterprise
Pricing Model Advertising
Ownership Public
AI-native platform that assesses and optimizes security controls across the stack.
AI/ML Security Services assess AI systems for adversarial attacks, model theft, prompt injection, and data integrity risks.
AI-native platform that assesses and optimizes security controls across the stack.
AI/LLM Security Assessment tests LLM applications and infrastructure for prompt injection, data poisoning, and model extraction risks.
AI/LLM Security Assessment tests LLM applications and infrastructure for prompt injection, data poisoning, and model extraction risks.
Security platform for protecting generative AI and LLM applications in enterprises.
Platform securing machine learning models and AI systems against attacks.
AI security agent that hunts intent-based vulnerabilities such as broken access control and business logic flaws.
AI/ML Security assesses machine learning pipelines, model architecture, and AI-based systems for security vulnerabilities.
Governed, citation-grounded assistant embedding Summarize, Analyze, and Create in public-safety workflows on Peregrine Foundation data.
Platform that discovers, monitors, and governs enterprise AI agents at runtime.
Continuous behavioral testing and remediation platform for AI agents and models.
One HIPAA-compliant gateway routing all LLM traffic with BAA coverage, audit logging, key management, and cost controls enforced automatically.
Unit21 product deploying AI agents across financial crime investigation workflows.
Security platform for protecting generative AI and LLM applications in enterprises.
Lakera Guard is a runtime security product for large language model applications.
Platform securing machine learning models and AI systems against attacks.
AI security agent that hunts intent-based vulnerabilities such as broken access control and business logic flaws.
AI/ML Security Services assess AI systems for adversarial attacks, model theft, prompt injection, and data integrity risks.
AI/ML Security assesses machine learning pipelines, model architecture, and AI-based systems for security vulnerabilities.
AI agent discovery and graph ontology inventory.
Governed, citation-grounded assistant embedding Summarize, Analyze, and Create in public-safety workflows on Peregrine Foundation data.
Platform that discovers, monitors, and governs enterprise AI agents at runtime.
Continuous behavioral testing and remediation platform for AI agents and models.
One HIPAA-compliant gateway routing all LLM traffic with BAA coverage, audit logging, key management, and cost controls enforced automatically.
Unit21 product deploying AI agents across financial crime investigation workflows.
