Skip to main content

Transparent pricing for AI teams

One record-based meter covers AI observability, evals, application performance, and infrastructure. $20 of free usage a month (10M records) on every plan, then $2 per million on Team and Growth.

Personal

Free, forever

Perfect for solo devs, and the agents building with them.

Logfire plan highlights

  • 10M telemetry records/month (hard-capped at $0)
  • Monitor agents, services, and hosts
  • 3 projects · 30-day retention
  • 1 seat + 2 read-only guests
  • Try Agent Sandboxes for free
See all 11 features

What you get

  • Monitor your agents

    Trace agent runs from prompt through tool calls with Pydantic AI, LangChain, OpenAI SDK, or anything OTel-instrumented.

  • Monitor your services

    Supports all OTel SDKs, so every service lands on one correlated timeline.

  • Monitor your hosts

    Kubernetes, hosts, databases, and more, with logs and host metrics with pre-built dashboards.

  • AI Gateway with BYOK

    Bring up to 3 provider credentials (no markup). Access built-in providers with a 5% markup fee (requires adding a credit card).

  • Fleet management (Beta)

    Configure your OpenTelemetry Collectors and SDKs from Logfire over OpAMP. Up to 3 SDK applications, 3 Collector policies and 20 connected Collectors. Learn more

Scale

  • Up to 10 million records

    Covers 10 million logs, spans, and metrics every month. This is a lot of telemetry for a solo developer! Once you hit the limit, ingestion pauses, so you can't owe us anything on the personal tier. Upgrade to Team for additional records.

  • 1 seat

    One admin seat (you).

  • 2 guests (read-only)

    A guest is a read-only user that can be invited directly to a Logfire project, but not to an organization. Guests are similar to outside collaborators in GitHub.

  • 3 projects

    Organize your apps.

  • 30-day data retention

    Query your traces, logs, and metrics for the last 30 days.

Support & compliance

  • Community support

    Get help from the Pydantic community on Slack, GitHub Discussions, and Stack Overflow. No SLA; response times depend on the community.

Team

$49/mo

For small teams shipping their first production workloads.

Logfire plan highlights

Everything in Personal, plus:

  • 10M records included · $2/M after, up to your spending cap
  • Unlimited agent, service, and host monitoring
  • AI output evaluations + human review
  • 5 seats included · 10 read-only guests
  • Agent Sandboxes from $49/mo
See all 13 features

What you get

  • $2 per million additional records

    10 million logs, spans, and metrics are free every month. Additional records are $2 per million, or set a hard price cap to lock in your bill.

  • Monitor unlimited agents

    No per-agent fees. Trace every agent run and only pay for the telemetry (records) you send.

  • Monitor unlimited services

    Supports all OTel SDKs, with every service on one correlated timeline with service inventory, service maps, and more. No per-service fees.

  • Monitor unlimited hosts

    Kubernetes, hosts, databases, and more, with logs and host metrics with pre-built dashboards and alerts.

  • $0/1000 Scores

    Evaluate your AI outputs for free, with no per-score charges.

  • Human Annotations

    Add human review to any trace or eval run, for RLHF training or manual quality checks.

  • AI Gateway with BYOK

    Bring up to 5 provider credentials (no markup). Access built-in providers with a 5% markup fee.

  • Fleet management (Beta)

    Configure your OpenTelemetry Collectors and SDKs from Logfire over OpAMP. Up to 5 SDK applications, 5 Collector policies and 100 connected Collectors. Learn more

Scale

  • Up to 12 seats (5 included)

    Admin or member seats for you & your core team. 5 seats included, $25/extra seat.

  • 10 guests (read-only)

    A guest is a read-only user that can be invited directly to a Logfire project, but not to an organization. Guests are similar to outside collaborators in GitHub.

  • 5 projects

    Organize your apps. Scope access per project.

Support & compliance

  • Spending cap

    Set a hard limit on your bill. Once you hit your cap, new telemetry is hidden from view, but we keep capturing it up to a generous threshold. To set a cap, email accounts@pydantic.dev. Learn more

  • Money-back guarantee

    Unsure after your first month? We'll refund you. And if you get hit with an unexpectedly large bill in your first month because of a spike in volume, we'll refund that too.

Growth

Most popular

$249/mo

For teams running production AI services core to their business.

Logfire plan highlights

Everything in Team, plus:

See all 10 features

What you get

  • AI Gateway with BYOK

    Bring up to 20 provider credentials (no markup). Access built-in providers with a 3% markup fee.

  • Fleet management (Beta)

    Configure your OpenTelemetry Collectors and SDKs from Logfire over OpAMP. Up to 50 SDK applications, 50 Collector policies and 1,000 connected Collectors. Learn more

  • Early access to features

    Try new Logfire capabilities before they ship broadly. Feedback shapes what lands and when.

Scale

  • All seats included, flat price

    No per-seat cost. Add your whole team.

  • Read-only guests for the whole company

    No cap on read-only viewers. Bring in stakeholders, clients, or your CTO.

  • Projects for every service

    No limits on services, environments, or experiments.

  • Up to 90-day retention

    Query your traces, logs, and metrics for the last 90 days, vs 30 on lower tiers.

Support & compliance

  • Priority support

    Faster response times when you need help.

  • Data deletion (GDPR)

    Self-serve deletion of users and trace data.

  • Boilerplate BAA (HIPAA)

    Ready-to-sign agreement for healthcare compliance.

Enterprise

Custom

For organizations where observability has to pass security review.

Logfire plan highlights

Everything in Growth, plus:

See all 13 features

What you get

  • Observability SLOs

    Define Service Level Objectives on your traces and metrics, with error budgets and burn-rate alerts. Includes LLM provider SLOs, which measure SLIs (latency, error rate) on the upstream LLM providers your enterprise services depend on.

  • Data Loss Prevention

    Policy-based redaction and blocking for sensitive data (PII, credentials, regulated content) before it reaches an LLM or lands in your logs. Requires Full AI Gateway access (add-on).

  • Discounted usage rates

    Volume-discounted rates negotiated against your usage commit.

  • Cloud, Dedicated, or Self-hosted

    Fully managed cloud, single-tenant infrastructure, or self-hosted on your own cluster.

  • Single sign-on: Okta, Microsoft Entra ID, Keycloak, SAML

    Bring your identity provider, with first-party integrations for the common IDPs and generic SAML for the rest.

  • SCIM group provisioning

    Discover users and provision group membership from your identity provider.

  • Full AI Gateway access (add-on)

    Unlimited BYO keys and built-in LLM providers, with enhanced Data Loss Prevention support on the LLM traffic path. Available as add-on.

Scale

  • Unlimited seats

    Add your whole organization, no per-seat cost.

  • Unlimited projects

    No cap on services, environments, or business units.

  • Fleet management at scale

    Unlimited SDK applications and Collector policies, and 10,000 connected Collectors on Cloud (more on request). Learn more

  • Multi-year retention

    Custom retention policies to meet your data and compliance requirements.

Add Agent Sandboxes to any Logfire plan

Agent Sandboxes Beta

Runs agent-written code in an isolated interpreter, powered by Pydantic Monty. Priced per invocation and billed on your Logfire plan.

Free

$0/mo

On every Logfire plan

50,000 invocations a month, free

Get started free
$1k beta credit

Pro

$49/mo

Add to a Personal or Team Logfire account.

$0.10 per 1,000 invocations

after 50,000 free each month

  • Included with Growth and Enterprise Cloud
Subscribe

Advanced

Custom

For high volumes and bespoke requirements.

  • Volume pricing
Contact us
Limit FreeProAdvanced
CPU time
50 ms per invocation
50 ms per invocation
Custom
Memory
20 MiB per session
20 MiB per session
Custom
Host calls
1,000 per session
1,000 per session
Custom
Session starts
300 every 5 minutes
300 every 5 minutes
Custom
Snapshot storage
2 GiB and 128 snapshots per project
2 GiB and 128 snapshots per project
Custom
Snapshot retention
7 days
7 days
Custom
Notes
Limits are fixed. An invocation that exceeds one fails.
Limits are fixed. An invocation that exceeds one fails.
Built around your volume and workloads. Contact us.

Concurrent sessions and projects with sandboxes follow your Logfire plan, whichever tier you choose: 4 sessions and 1 project on Personal, no session limit and 3 projects on Team, Growth and Enterprise.

Cost calculator

Usage rate
$2/M
Ingested Logs/Spans/Metrics
10M
Free allowance
- 10M
Payable
= 0M
Calculation
$2/M x 0M = $0
Usage total
$0
Base fee (5 seats included)
$49
Extra seats (0 × $25)
$0
Agent Sandboxes Free
$0
Within 50,000 free invocations
Monthly total
$49
Get Started

Detailed comparison

Feature Personal Team Growth Enterprise
Usage
Base cost Free $49/mo $249/mo Custom
Seats 1 5 included, $25/extra seat Unlimited Unlimited
Guests (read-only) 2 10 Unlimited Unlimited
Projects 3 5 Unlimited Unlimited
Logs/Spans/Metrics included 10M 10M 10M Custom
Additional usage Not allowed $2/M $2/M Custom
Price cap $0 Included Included Included
Payment method Not included Credit card Credit card Invoice
AI Gateway
Availability Included Included Included Add-on
Bring your own provider credentials (BYOK) Up to 3 Up to 5 Up to 20 Unlimited
BYOK markup 0% 0% 0% 0%
Built-in providers markup 5% 5% 3% 3% (not on Self-hosted)*
Fleet management (Beta)
Availability Included Included Included Included
SDK applications 3 5 50 Unlimited
Collector policies 3 5 50 Unlimited
Connected Collectors 20 100 1,000 10,000 on Cloud, custom on Dedicated, unlimited on Self-hosted*
Agent Sandboxes
Free sandboxes Included Included Included Included (Cloud)*
Pro $49/mo $49/mo Included Included (Cloud)*
Advanced Custom Custom Custom Custom
Deployment
Hosting Logfire Cloud, ready in minutes Logfire Cloud, ready in minutes Logfire Cloud, ready in minutes Cloud, Dedicated, or Self-hosted*
Infrastructure Fully managed Fully managed Fully managed Managed or your own*
Scalability Managed scaling Managed scaling Managed scaling Managed or autoscaling*
Data
EU or US Data region Included Included Included Included Cloud, any GCP region on Dedicated*
On-prem Not included Not included Not included Included Self-hosted*
Data retention 30 Days 30 Days Up to 90 Days Longer than 90 days or custom*
Self-service data deletion Not included Not included Included Included *
Query API requests/day ≤500 requests ≤500 requests ≤5,000 requests Custom*
Public API Not included Not included Included Included *
Custom roles/permissions Not included Not included Not included Included *
Support
Public Slack channel Included Included Included Included
Priority support Not included Not included Included 24/7
Setup assistance Not included Not included Not included Self-hosted*
Dedicated Slack channel Not included Not included Not included Included
Compliance & scale
Single tenant infra Not included Not included Not included Included Dedicated, negotiable on Cloud*
SSO (Okta, Entra ID, Auth0, LDAP, Google Workspace) Not included Not included Not included Included
SCIM group provisioning Not included Not included Not included Included
Audit log API Not included Not included Not included Included
DPA (GDPR) Standard Standard Standard Custom
BAA (HIPAA) Not included Not included Boilerplate Custom
SOC 2 Type 2 Included Included Included Included Cloud and Dedicated*
SLA Not included Not included Not included Included
CMEK Not included Not included Not included Included Dedicated*
VPC Peering Not included Not included Not included Optional on Dedicated*

* Varies by Enterprise deployment: Cloud, Dedicated or Self-hosted. Compare deployment options

We offer financial support for Not-for-profits, Academics, Community groups, and pre-funding Start-ups.

Apply now (opens in a new tab)

Logfire pricing FAQs

What happens if I use more than the free allowance?

For all plans we'll send you emails and in-app alerts on your level of use. On Personal, if you go well over the free allowance, we'll eventually stop ingesting new data, and you might see warnings in stdout/stderr about 4XX responses from the Logfire API. For Team, Growth and Enterprise customers you'll be charged at the plan specific rate for your additional usage.

Can I use the Personal plan in production?

Yes! The Personal plan is designed for you to test Logfire or use it with side projects. You can safely use it in production if you accept that ingestion pauses once you hit the 10 million records per month cap, so you can't owe us anything on Personal.

What is the price cap?

We offer a spending cap to all paying customers. Once you hit your cap, new telemetry is hidden from view, but we keep capturing it up to a generous threshold. Raise your cap anytime to unlock your data, so you'll never lose visibility when you need it most. To set a spending cap, email us at accounts@pydantic.dev.

How long is my data stored in Logfire?

30 days on Personal and Team, up to 90 days on Growth. Enterprise customers can configure custom retention (unlimited available).

What counts toward the Fleet management limits?

Limits count across your whole organization, and the comparison above lists them for each plan. Collector policies include Kubernetes (Operator) policies; SDK policies do not count. Need more? Talk to us. Read more about Fleet management.

Will Pydantic remain free?

The Pydantic data validation library is (always has been, always will be) completely free and permissively licensed under the MIT license. Prices displayed on this page apply to Pydantic Logfire, only.

I'm currently on the Free tier. Will I have to pay from now on?

If you're the only user on an account with up to 3 projects and up to 10 million logs/spans/metrics, nothing changes except your plan name (to Personal). Your usage is still free. If you exceed these limits or have additional users, you'll be asked to upgrade. If you don't upgrade, data ingestion will eventually be suspended and orgs other than your Personal account will be disabled.

I'm currently on the free tier and have exceeded the allowance of 10 million logs/span/metrics. Can I pay for the extra usage ad-hoc?

Ad-hoc payments for additional records are not available on the Personal plan. If your usage exceeds the 10 million record allowance, you'll need to upgrade your plan.

What counts as an Agent Sandboxes invocation?

One feed of code to a sandbox, however many host callbacks it suspends for. Each one is capped at 50 ms of CPU and runs in a session capped at 20 MiB of memory, so on Pro the invocation price is the whole bill.

Which Agent Sandboxes tier do I need?

Free is for trying the product. Pro runs production workloads that fit fixed limits of 50 ms CPU per invocation and 20 MiB of memory per session. Advanced is for high volumes or workloads that need more than those limits: we set its limits and price with you, so contact us.

What does “included” mean for a sandbox tier on a Logfire plan?

It waives the tier's monthly fee. Pro still includes 50,000 free invocations a month, and invocations past those are billed at $0.10 per 1,000.

Can I buy Agent Sandboxes without Logfire, or self-host them?

Sandboxes are part of Logfire and run in Logfire Cloud. Personal costs nothing, so Pro on Personal costs the $49 sandbox fee plus usage. Self-hosting is planned as part of Logfire's self-hosted Enterprise deployment, early next year.

Pricing terms

Record
A record is one log, span, or metric sent to Logfire.
Log
A log is the record of an event. A flat data structure with timestamp, level (debug, info, warning, error, critical), and a message. Although Logfire can receive logs, we recommend you use it to store high-value events from your application, such as spans and traces.
Span
A span is the building block of a trace, represented by a single row in the Logfire live view. Spans have start and end times, attributes (key-value metadata), and can nest under parent spans. Multiple (nested) spans form a trace, which represents a complete request flow or operation.
Metric
A metric is a collection of datapoints at intervals. When stored with timestamps, metrics become timeseries. While traces show what happened in a request and logs capture single events, metrics show trends across your system.
Guest
A guest is a read-only user that can be invited directly to a Logfire project, but not to an organization. Guests are similar to outside collaborators in GitHub.
Price cap
Set a hard limit on your bill. Once you hit your cap, new telemetry is hidden from view, but we keep capturing it up to a generous threshold. Raise your cap anytime to unlock your data. To set a cap, email accounts@pydantic.dev.
Agent Sandboxes beta credit
Get $1,000 of Agent Sandboxes credit on us while we’re in beta. It covers your Pro fee and usage, which is around 9 million invocations at $0.10 per 1,000.

Request received

One more step: book your meeting

We're sending a booking link to the work email you entered.

  1. Check your inbox

    Look for an email from Pydantic.

  2. Choose a time

    Open the booking link and pick a slot.

Your meeting isn't booked until you choose a time.

No email after a few minutes? Check your spam folder or contact sales@pydantic.dev.

Book a call

Tell us a bit about you and we'll be in touch to find a time.

By submitting, you acknowledge that we will process the information in this form as described in our Privacy Policy.

Get started with Logfire for free

Sign up today and see your first trace in under 2 minutes

01 Install

$ uv add logfire

Resolved 7 packages in 128ms
Installed 7 packages in 89ms
+ logfire

02 Instrument

import logfire

logfire.configure()
logfire.info('Hello {name}', name='world')

Free up to 10m records per month No credit card needed