PRICING

Pricing that scales with you

From your first prototype to production at planetary scale. Start free, then pay only for what you use — no seats, no surprises, no lock-in.

Save 20% with annual billing

Free

$0/mo

For prototypes and side projects. No credit card required.

Start for free
  • 100K tokens / month included
  • 1 project, 1 environment
  • Community models + shared inference
  • Basic tracing (7-day retention)
  • Community support

Pro

Most popular
$49/mo

For teams shipping their first agents to production.

Start free trial
  • 10M tokens / month included
  • Unlimited projects, 3 environments
  • Dedicated inference routing
  • Full tracing + guardrails (30-day retention)
  • Evaluations with rubric judges
  • Email support, 1-business-day SLA

Scale

$249/mo

For high-volume workloads that need headroom and priority.

Start free trial
  • 100M tokens / month included
  • Unlimited environments
  • Warm capacity in every region
  • Advanced guardrails + policy-as-code
  • 90-day trace + eval retention
  • Priority support, 4-hour SLA

Enterprise

Custom

For organizations with security, compliance, and scale needs.

Talk to sales
  • Volume token pricing + committed spend
  • Dedicated + private-region capacity
  • SSO / SCIM, audit logs, RBAC
  • Custom guardrail policies + BYO models
  • Unlimited retention + data residency
  • 24/7 support, dedicated solutions engineer

COMPARE PLANS

Every feature, side by side

INCLUDED EVERYWHERE

Every plan includes

Features Free Pro Scale Enterprise
Usage & limits
Tokens included / month100K10M100MCustom
Projects1UnlimitedUnlimitedUnlimited
Environments13UnlimitedUnlimited
Overage rate (per 1M tokens)$0.60$0.45Volume
Inference & routing
Shared inference
Dedicated routing
Warm capacity, all regions
Private-region capacity
Quality & safety
Distributed tracing
Guardrails (policy-as-code)
Evaluations + rubric judges
Trace + eval retention7 days30 days90 daysUnlimited
Security & governance
SSO / SCIM
Audit logs + RBAC
Data residency
Support
Support channelCommunityEmailPriority24/7 dedicated
Response SLA1 business day4 hoursCustom
  • Typed SDK for Python & TypeScript
  • 40+ native model & data connectors
  • Streaming, retries & structured outputs
  • Usage dashboards & spend alerts
  • SOC 2 Type II infrastructure
  • No per-seat pricing, ever

USAGE-BASED

Only pay for what you use

Beyond your plan's included allowance, every workload is metered per unit. No seats, no minimums — the meter stops when your traffic does.

Inference tokens

Metered per million tokens across all models. Free up to your plan's monthly allowance, then billed by volume tier.

$0.45/ 1M tokens

Agent compute

Metered per GB-hour of agent runtime for long-running and scheduled workloads. Free up to 50 GB-hours each month.

$0.018/ GB-hour

Vector storage

Metered per GB-month of stored embeddings and retrieved context. Free up to 5 GB each month.

$0.10/ GB-month

FAQ

Pricing questions, answered

GET STARTED

Start free, scale when you are ready

Spin up your first agent in minutes on the Free plan — no credit card, no seats. Talk to us when you need dedicated capacity or enterprise controls.