Langfuse launches AI Gateway ยท Read more

Pricing

Deploy Langfuse OSS today. Upgrade to Enterprise at any time.

Open Source
Self-host all core Langfuse features for free without any limitations.
  • MIT License
  • All core platform features and APIs (observability, evaluation, prompt management, datasets, etc.)
  • Scalability of Langfuse Cloud
  • Deployment docs & Helm chart
  • Enterprise SSO and RBAC
  • Community support
Enterprise
Dedicated Langfuse deployment with enterprise capabilities and support.
  • All Open Source features plus management APIs, project-level RBAC, data retention policies, and audit logs
  • Bundled with ClickHouse Cloud, ClickHouse BYOC, or ClickHouse Private
  • Langfuse pricing is additive to your ClickHouse commercial plan
  • Named lead support engineer for deployment and hosting guidance
  • Solutions architect support during evaluation and rollout
  • Direct access to the product team for feedback
  • SOC 2 Type II and ISO 27001 reports
  • Support SLA
  • Billing via AWS Marketplace or invoice
Government
Dedicated Langfuse deployment for government and regulated industries.
Custom Pricing
  • Everything in Enterprise
  • FIPS 140-3 validated cryptographic modules operated in approved mode
  • Iron Bank-hardened UBI 9 base images
  • Air-gapped deployment

Feature comparison

Open Source

Self-host all core Langfuse features for free without any limitations.

LLM Application & Agent Tracing
Traces and Graphs (Agents)
Session Tracking (Chats/Threads)
User Tracking
Token and Cost Tracking
Native Frameworks Integrations
SDKs (Python, JavaScript)
OpenTelemetry (Java, Go, custom)
Proxy-based Logging (via Langfuse AI Gateway, LiteLLM, etc.)
Custom via API
Included usage
Unlimited
Multi-modal
External media storage (S3)
Langfuse AI
Halo (in-app agent)
MCP
CLI
Prompt Management
Prompt Versioning
Prompt Fetching
Unlimited
Prompt Release Management
Prompt Composability
Prompt Caching (server and client)
Playground
Prompt Experiments
Webhooks & Slack
Protected Deployment Labels
Evaluation (online and offline)
Datasets
Experiments via SDK
Experiments via UI
Evaluation Scores (custom)
User Feedback Tracking
External Evaluation Pipelines
LLM-as-judge evaluators
Human Annotation
Human Annotation Queues
Metrics
Custom Dashboards
Alerts
Langfuse v4+
Collaboration
Projects
Unlimited
Users
Unlimited
AI Gateway
AI Gateway
API
Extensive Public API
Metrics & Observations APIs (v2)
Langfuse v4+
Exports
Batch Export via UI
PostHog Integration
Mixpanel Integration
Scheduled Batch Export to Blob Storage
Deployment
ClickHouse deployment model
Self-managed ClickHouse OSS
Deployment templates
Local (Docker Compose)
Kubernetes (Helm)
AWS (Terraform)
Azure (Terraform)
GCP (Terraform)
Support
Ask AI
Community (GitHub)
In-app support
Private Slack / MS Teams channel
Named Lead Support Engineer
Onboarding & Architectural guidance
Solutions architect support
Product team feedback channel
Support SLA
Security
Sign in with Google, AzureAD, GitHub
Organization-level RBAC
Enterprise SSO (e.g. Okta, AzureAD/EntraID)
SSO enforcement
Client-side data masking
Server-side data masking
Project-level RBAC
Data retention management
SCIM API for automated user provisioning
Organization Creators
UI Customization
Audit Logs
IAM role authentication for Amazon S3
Admin API (project management, SCIM)
Instance Management API
FIPS 140-3 validated cryptographic modules operated in approved mode
Iron Bank-hardened UBI 9 base images
Billing
Subscription management
Payment methods
Contract duration
Billing via AWS Marketplace
Compliance
Contracts
SOC2 Type II & ISO27001 reports
InfoSec/legal reviews

Enterprise

Dedicated Langfuse deployment with enterprise capabilities and support.

LLM Application & Agent Tracing
Traces and Graphs (Agents)
Session Tracking (Chats/Threads)
User Tracking
Token and Cost Tracking
Native Frameworks Integrations
SDKs (Python, JavaScript)
OpenTelemetry (Java, Go, custom)
Proxy-based Logging (via Langfuse AI Gateway, LiteLLM, etc.)
Custom via API
Included usage
Unlimited
Multi-modal
External media storage (S3)
Langfuse AI
Halo (in-app agent)
MCP
CLI
Prompt Management
Prompt Versioning
Prompt Fetching
Unlimited
Prompt Release Management
Prompt Composability
Prompt Caching (server and client)
Playground
Prompt Experiments
Webhooks & Slack
Protected Deployment Labels
Evaluation (online and offline)
Datasets
Experiments via SDK
Experiments via UI
Evaluation Scores (custom)
User Feedback Tracking
External Evaluation Pipelines
LLM-as-judge evaluators
Human Annotation
Human Annotation Queues
Metrics
Custom Dashboards
Alerts
Langfuse v4+
Collaboration
Projects
Unlimited
Users
Unlimited
AI Gateway
AI Gateway
API
Extensive Public API
Metrics & Observations APIs (v2)
Langfuse v4+
Exports
Batch Export via UI
PostHog Integration
Mixpanel Integration
Scheduled Batch Export to Blob Storage
Deployment
ClickHouse deployment model
Bundled: ClickHouse Cloud / BYOC / Private
Deployment templates
Local (Docker Compose)
Kubernetes (Helm)
AWS (Terraform)
Azure (Terraform)
GCP (Terraform)
Support
Ask AI
Community (GitHub)
In-app support
Private Slack / MS Teams channel
Named Lead Support Engineer
Onboarding & Architectural guidance
Solutions architect support
Product team feedback channel
Support SLA
Security
Sign in with Google, AzureAD, GitHub
Organization-level RBAC
Enterprise SSO (e.g. Okta, AzureAD/EntraID)
SSO enforcement
Client-side data masking
Server-side data masking
Project-level RBAC
Data retention management
SCIM API for automated user provisioning
Organization Creators
UI Customization
Audit Logs
IAM role authentication for Amazon S3
Admin API (project management, SCIM)
Instance Management API
FIPS 140-3 validated cryptographic modules operated in approved mode
Iron Bank-hardened UBI 9 base images
Billing
Subscription management
Payment methods
Contract duration
Billing via AWS Marketplace
Compliance
Contracts
SOC2 Type II & ISO27001 reports
InfoSec/legal reviews

Government

Dedicated Langfuse deployment for government and regulated industries.

LLM Application & Agent Tracing
Traces and Graphs (Agents)
Session Tracking (Chats/Threads)
User Tracking
Token and Cost Tracking
Native Frameworks Integrations
SDKs (Python, JavaScript)
OpenTelemetry (Java, Go, custom)
Proxy-based Logging (via Langfuse AI Gateway, LiteLLM, etc.)
Custom via API
Included usage
Unlimited
Multi-modal
External media storage (S3)
Langfuse AI
Halo (in-app agent)
MCP
CLI
Prompt Management
Prompt Versioning
Prompt Fetching
Unlimited
Prompt Release Management
Prompt Composability
Prompt Caching (server and client)
Playground
Prompt Experiments
Webhooks & Slack
Protected Deployment Labels
Evaluation (online and offline)
Datasets
Experiments via SDK
Experiments via UI
Evaluation Scores (custom)
User Feedback Tracking
External Evaluation Pipelines
LLM-as-judge evaluators
Human Annotation
Human Annotation Queues
Metrics
Custom Dashboards
Alerts
Langfuse v4+
Collaboration
Projects
Unlimited
Users
Unlimited
AI Gateway
AI Gateway
API
Extensive Public API
Metrics & Observations APIs (v2)
Langfuse v4+
Exports
Batch Export via UI
PostHog Integration
Mixpanel Integration
Scheduled Batch Export to Blob Storage
Deployment
ClickHouse deployment model
ClickHouse Government
Deployment templates
Local (Docker Compose)
Kubernetes (Helm)
AWS (Terraform)
Azure (Terraform)
GCP (Terraform)
Support
Ask AI
Community (GitHub)
In-app support
Private Slack / MS Teams channel
Named Lead Support Engineer
Onboarding & Architectural guidance
Solutions architect support
Product team feedback channel
Support SLA
Security
Sign in with Google, AzureAD, GitHub
Organization-level RBAC
Enterprise SSO (e.g. Okta, AzureAD/EntraID)
SSO enforcement
Client-side data masking
Server-side data masking
Project-level RBAC
Data retention management
SCIM API for automated user provisioning
Organization Creators
UI Customization
Audit Logs
IAM role authentication for Amazon S3
Admin API (project management, SCIM)
Instance Management API
FIPS 140-3 validated cryptographic modules operated in approved mode
Iron Bank-hardened UBI 9 base images
Billing
Subscription management
Payment methods
Contract duration
Billing via AWS Marketplace
Compliance
Contracts
SOC2 Type II & ISO27001 reports
InfoSec/legal reviews

Discounts

Early-stage startups
50% off, first year
Research / Students
Up to 100% off, limits apply
Non-profits
USD 199 in credits / month
Open-source projects
USD 300 in credits / month, first year

Reach out to support@langfuse.com to apply for a discount. We want all startups, educational users, non-profits and open source projects to build with Langfuse and are happy to work with you to make that happen.

FAQ

You can view the public example project or sign up for a free account to try Langfuse with your own data. The Hobby plan is completely free and does not require a credit card.

Yes, Langfuse is open source and you can self-host Langfuse for free. Use docker compose to run Langfuse locally, or use one of our templates to self-host Langfuse in production on Kubernetes. Check out the self-hosting documentation to learn more.

A billable unit in Langfuse is any tracing data point you send to our platform โ€” this includes the trace (a complete application interaction), observations (individual steps within a trace: Spans, Events and Generations), and scores (evaluations of your AI outputs). For a detailed explanation and an example, see our Langfuse Billable Units docs.

Our graduated pricing means you pay different rates for different volume tiers. The first 100k units are included in paid plans, then you pay $8/100k for units 100kโ€“1M, $7/100k for 1Mโ€“10M units, $6.5/100k for 10Mโ€“50M units, and $6/100k for 50M+ units. This ensures you get better rates as you scale up your usage. Use the pricing calculator to estimate your bill.

Data access is set by your plan and determines how many days of historic data you can view (Hobby: 30 days, Core: 90 days, Pro and Enterprise: 3 years). Older data is still stored and becomes accessible again when you upgrade. Data retention is a separate, opt-in feature on Pro and Enterprise that permanently deletes data after a custom period. See data access vs. data retention.

The primary way to reduce your Langfuse Cloud bill is to reduce the number of billable units that you ingest. We have summarized how this can be done here. Additionally, with our new graduated pricing model, you automatically get lower rates per 100k units as your volume increases.

On a representative workload, Langfuse Cloud is cheaper than LangSmith, Braintrust, Arize, and Datadog Agent Observability. Open the editable pricing comparison sheet to fill in your own assumptions and check the numbers for your workload. For product-by-product differences, see the comparison pages. Langfuse is also open source, based on OpenTelemetry for interoperability, self-hostable at the same scale as Langfuse Cloud, and model- and framework-agnostic.

You get one bill each month. We charge your Core, Pro, or Team plan at the start of the month. We charge for your usage at the end of the month. The bill you get at the start of the month shows two things: the plan cost for the new month and the usage from last month.

Yearly commitments are always available for Enterprise; Core, Pro, and Teams customers can also choose a yearly commitment if they meet the minimum contract value threshold.

A Langfuse Cloud Credit is a prepaid unit of credit, equal to one (1) US dollar, to be applied to a customer's use of Langfuse Cloud. Credits are drawn down according to the then-current pricing at langfuse.com/pricing. Langfuse Cloud Credits only apply to customers with a committed spend contract for Langfuse Cloud. If you have questions about your credit balance, drawdown schedule, or committed spend contract, please contact your account team. If you use Langfuse Cloud on a Pay-as-you-Go basis, Langfuse Cloud Credits do not apply to you.

Yes, you can configure spend alerts to receive email notifications when your organization's spending exceeds predefined monetary thresholds. This helps you monitor costs and take action before unexpected charges occur. Navigate to your organization settings and the Billing tab to configure spend alerts. Learn more in our spend alerts documentation.

You can manage your subscription through the organization settings in Langfuse Cloud or by using this Customer Portal.

Yes. Self-hosted Enterprise can be billed via AWS Marketplace or invoice. Talk to us to request a private offer.

Yes, we offer customized contracts for Langfuse Enterprise customers with a yearly commitment. Please contact sales for more details. The default plans are affordable as they are designed to be self-serve on our standard terms.

Langfuse Cloud is hosted on AWS and data is stored in the US, EU, or Japan depending on your selection. Enterprise customers can also choose a HIPAA-compliant US region. See our security and privacy documentation for more details.

We publish almost all of our security documentation and controls. Please refer to our security documentation for more details.