About Langfuse
Langfuse is an open source LLM engineering platform offering observability, evaluation, prompt management, and metrics for debugging and improving LLM applications. It serves teams building complex LLM apps and provides end-to-end visibility to observe, test, and iterate model behavior.
Key Features
When you’re shipping LLM-powered flows, you need visibility into traces, evaluation data, and prompt histories to debug quickly and iterate confidently:
Traces
Capture complete traces of LLM applications/agents; inspect failures and build eval datasets. Based on OpenTelemetry with support for all popular LLM/agent libraries.
Evals
Run, benchmark, and refine model responses against curated evaluation datasets to improve quality and consistency.
Prompt Management
Centralize prompts across models, versions, and deployments with versioning and testing support.
Metrics
Collect and analyze latency, throughput, and success rates across the LLM stack for performance visibility.
Summary
Best for ML/AI engineering teams, product teams shipping LLM workflows, and platform/DevOps squads that need end-to-end observability.
Pricing
View pricingHobby
Free / Monthly
- All platform features (with limits)
- 50k units / month included
- 30 days data access
- 2 users
- Community support via GitHub
Core
$29 / Monthly
- Everything in Hobby
- 100k units / month included, additional: $8/100k units. Lower with volume
- 90 days data access
- Unlimited users
- In-app support
Pro
$199 / Monthly
- Everything in Core
- 100k units / month included, additional: $8/100k units. Lower with volume
- Unlimited data access
- Data retention management
- Unlimited annotation queues
- High rate limits
- SOC2 & ISO27001 reports, BAA available (HIPAA)
- Prioritized in-app support
Teams Add-on
$300 / Monthly
- Enterprise SSO (e.g. Okta)
- SSO enforcement
- Fine-grained RBAC
- Support via Dedicated Slack Channel
Enterprise
$2499 / Monthly
- Everything in Pro + Teams
- 100k units / month included, additional: $8/100k units. Lower with volume
- Audit Logs
- SCIM API
- Custom rate limits
- Uptime SLA
- Support SLA
- Dedicated support engineer
