Agent engineering built for EU sovereignty and regulation: run LangSmith in our EU cloud, your cloud, or fully self-hosted, with the audit trail and controls the EU AI Act expects.
Build agents fast with or without code. Choose your model. Own your harness.
Validate behavior before you ship. Turn traces into realistic datasets and eval every step.
Ship agents on infra built for long-running tasks, persistent state, and secure execution.
Catch issues in production with online evals, dashboards, and alerts built for agents.
Compound agent intelligence through a feedback loop you own.
Enforce cost limits and detect sensitive data before damage happens.
LangSmith cloud stores data in secure infrastructure. When using LangSmith hosted at smith.langchain.com, data is stored in GCP us-central-1 or europe-west4. If you’re on the Enterprise plan, we can deliver LangSmith to run on your kubernetes cluster in AWS, GCP, or Azure so that data never leaves your environment. For more information, check out our documentation. For teams with compliance requirements, self-hosted and BYOC options let you control where your data lives.
LangSmith is SOC 2 Type 2 and ISO 27001 certified, GDPR compliant, and HIPAA compliant. Audit reports, penetration test summaries, and a pre-signed DPA are available through the Trust Center (trust.langchain.com).
No. LangChain commits that it will not train on your data, and you own all rights to it. Retention windows are configurable and PII masking can hide sensitive fields before they reach LangSmith servers.
No. LangSmith is framework-agnostic and built on open standards, so your AI strategy is never tied to a single stack. You can trace langgraph, langchain, or deepagents, or integrate any stack via the Python, TypeScript, Go, and Java SDKs or OpenTelemetry. This enables users to swap models or frameworks without losing your observability, evaluations, or history.
Enterprise plans are tailored to your deployment model, trace volume, and support needs, with volume commitments, admin-level cost controls, and priority support with accelerated response-time SLAs. Talk to sales for a quote.
No. The LangSmith SDK uses an async callback handler that sends traces to a distributed collector. Your application performance is never impacted. If LangSmith experiences an incident, your agent keeps running normally.
Start building with an open-source coding agent you can customize, inspect, and control.
