Openlayer Guardrails

Overview

Guardrails can block, redact, escalate or route policy violations for review as inputs and outputs move through an application. Teams can tailor policies around sensitive-data categories and topics an assistant should avoid, making the product more adaptable than a fixed set of filters.

Its controls target prompt injection, jailbreak attempts, PII and PHI exposure, and harmful or toxic content. Openlayer says it detects and blocks more than 50 PII and PHI entity types inline. Each decision can carry a timestamp, system version, owner, supporting trace and reason for the block, which gives engineering and compliance teams a record to investigate and explain behavior.

Key features

Guardrails are commonly enforced through the Gateway, but can also be applied directly through the SDK. Openlayer connects through SDKs, framework integrations, a network-level gateway, CLI and REST API; the Gateway is designed to work across LLM providers. That range gives teams more than one integration path, though the best fit depends on where they want enforcement to sit in their stack.

Openlayer says guardrail activity feeds into compliance evidence and names the EU AI Act, ISO/IEC 42001, NIST AI RMF and OSFI E-23. This can help teams assemble evidence around controls, but the product's stated role is evidence support, not a substitute for an organization's compliance program.

Pricing

Basic: Free. It includes one member, five projects, one inference pipeline per project, 20,000 inference logs per month, 20 tests per project, one workspace, three months of data retention and Discord community support. The free tier is useful for a small evaluation or modest deployment, but the one-seat limit, project and pipeline caps, log quota and short retention window constrain broader team use.

Enterprise: custom pricing, billed Custom. It includes unlimited members and projects, custom inference pipelines and logs, unlimited tests, custom data retention, SAML SSO, on-premise deployment, white-glove onboarding and direct Slack access to Openlayer's team. This is the better fit for larger or regulated deployments that need identity controls, dedicated support or on-premise operation; teams must obtain a custom quote.

Platforms

Openlayer lists API, self-hosted and web platforms, with a hybrid deployment model. It supports deployment in a customer's VPC and air-gapped deployments; on-premise deployment is available on Enterprise. Openlayer states that it has SOC 2 Type II certification and GDPR compliance, supports HIPAA-aligned use cases, and encrypts data at rest with AES-256 and in transit with TLS 1.2 or higher.

Who it's for

Engineering teams that need configurable runtime checks across AI inputs and outputs are the clearest fit, particularly when security and compliance teams need decision-level records. Openlayer identifies engineering, AI, security, compliance and finance teams as users, including in healthcare and financial services. Basic is best for a narrow pilot; the Enterprise plan aligns better with organizations that need expanded capacity, SSO, custom retention or on-premise deployment.

Pros and cons

• Pros: Multiple enforcement paths, including a provider-agnostic Gateway and SDK, give teams flexibility in how they connect checks to applications.
• Pros: Configurable policies and inline detection across more than 50 PII and PHI entity types address both sensitive-data controls and topic restrictions.
• Pros: Decision records with traces and reasons can support investigations and compliance evidence.
• Cons: Basic is limited to one member, five projects and 20,000 inference logs per month, which can make it a poor fit for a growing team or busy production workload.
• Cons: Three months of Basic data retention may be too short for teams that need longer review windows.
• Cons: Enterprise pricing is custom, so buyers cannot compare its cost from a published rate.

Alternatives

For a broader shortlist, see LLM Security Tools and AI governance tools.

Prompt Inspector is another freemium option, with a free plan that includes 1,000 detections per month, REST API, SDK and MCP; choose it when that allowance and its listed integrations suit the workload. PromptGuard offers a free plan with 20,000 scans a month, one API key, one project and 24-hour log retention, plus a Team plan at 19.00 USD per month; it is worth considering when those terms fit better. GLACIS has a free account with no card required and setup of about a minute, making it an option for a quick, no-cost start.

Lasso AI Security Platform is a paid API, extension and web option. WitnessAI is a paid API, web and Windows offering with custom enterprise pricing. Prisma AIRS AI Gateway uses paid, consumption-based licensing through Software NGFW credits tied to token usage. SecureAI Guard has no free plan and a Startup plan at 2999.00 USD per month. GuardionAI is a paid option with a free trial and custom-contract Enterprise pricing.

Verdict

Openlayer Guardrails is a strong fit for teams that want configurable runtime enforcement, multiple integration paths and decision records that can contribute to compliance evidence. Its free Basic tier is a practical starting point, but its one-member cap, log quota and three-month retention make it a limited production choice. Look elsewhere if those limits are a mismatch or if custom Enterprise pricing is not workable.