Skip to content

Preview pricing. Final pricing at launch.

Priced on the failures you find.

You are not paying for compute. You are paying for review time that lands on real failures. The inference engine is billed per run; storing flagged runs in Hosted Langfuse is billed by size.

Plans

Free

For trying it on one agent.

$0per month

Inference engine
10,000 runs / month
Hosted Langfuse
1 GB storedincluded
  • 1 project, 1 agent
  • Send flagged runs to your own backend
  • Community support
Join the waitlist

Pay as you go

For teams with agents in production.

$1.00per 1,000 runs checked

Inference engine
$1.00 / 1,000 runsfirst 10,000 free
Over 1M runs / month
$0.50 / 1,000 runs
Hosted Langfuse
$0.50 / GB-monthfirst 1 GB free
  • Unlimited projects and agents
  • Quality alerts to Slack and PagerDuty
  • Spend caps across both meters
  • Management API
  • Email support
Join the waitlist

Enterprise

For high volume and strict requirements.

Customtailored to your volume

Inference engine
Volume pricingbelow $0.50 / 1,000 runs
Hosted Langfuse
Custom retentionand dedicated storage
  • Dedicated deployment
  • SSO
  • SLA
  • Priority support
Contact sales

Why this price

More of every review lands on a real failure.

Reviewer time is the expensive part of agent quality. Tailwise Cloud does not try to be the cheapest way to score a run. It competes on how many of your team's reviews find something worth fixing.

  • 50

    failures found in 500 random reviews

  • 237

    failures found in the same 500 reviews, using Tailwise

  • 4.7x

    more failures per review, measured on τ-bench with action checks

How billing works

Two meters, one bill.

  • Inference engineper run checked

    Every agent run that reaches Tailwise Cloud is checked once. Steps inside a run are not counted separately.

  • Hosted Langfuse storageper GB-month

    Flagged runs we store for you, measured by size and averaged over the month. Retention controls how long runs stay, and so how much you store.

Using your own Langfuse, Phoenix or Datadog? Storage is $0. Your backend, your bill.

Estimate

What would this cost you?

Runs per day

20,000

Average run size
Retention
Inference engine
$590.00
Storage(1.50 GB)
$0.25

Estimated per month

$590.25

Inference engine: 20,000 × 30 = 600,000 runs − 10,000 free = 590,000 billable → 590,000 × $1.00 / 1,000 = $590.00

Storage: 20,000 × 5% stored × 50 KB × 30 days = 1.50 GB − 1 GB free = 0.50 GB × $0.50 = $0.25

Pricing FAQ

Billing questions.

Early access

Stop finding out from your customers.

Join the waitlist for early access. Running agents at volume today? Talk to us.

Need volume pricing or a dedicated deployment? Contact sales