Pricing

Start free. Pay as your traffic grows.

Every plan gets the same models and the same speed. Higher tiers raise your limits and add the controls a security review asks for.

Feature
Free
$0 /mo

No card needed

Prototype with every fast model.

Start building
  • 20 requests per minute
  • 1M tokens a month
  • Fast and embedding models
  • Community support
Most popular
Pro
$40 /mo

Billed monthly

For apps with real users in production.

Start free trial
  • 500 requests per minute
  • Reasoning models
  • Batch API at half price
  • 5 team seats
Scale
$400 /mo

Billed monthly

High volume, tuned to your own data.

Upgrade to Scale
  • 5,000 requests per minute
  • Fine-tuning and evals
  • Zero data retention
  • Priority support
Enterprise
Custom

Annual contract

Private deployments and custom terms.

Contact sales
  • Private model deployments
  • SSO, SCIM and audit logs
  • Pin data to a region
  • 99.99% uptime SLA
Usage
Requests per minuteSustained rate across all models. Short bursts up to 2× are allowed.
20
500
5,000
Custom
Tokens per month
1M
50M
1B
Unlimited
Context windowThe most tokens a single request can read, prompt and files included.
32K
200K
1M
1M
Batch API
Not included
Included
Included
Included
Prompt caching
Not included
Included
Included
Included
Models
Fast models
Included
Included
Included
Included
Reasoning models
Not included
Included
Included
Included
Embeddings and rerank
Included
Included
Included
Included
Fine-tuningTrain on your own examples and serve the result at the base model's price.
Not included
Not included
Included
Included
Private deployments
Not included
Not included
Not included
Included
Collaboration
Team seats
1
5
25
Unlimited
Shared projects
Not included
Included
Included
Included
Usage alerts
Included
Included
Included
Included
Role-based access
Not included
Not included
Included
Included
Security
SSO (SAML, OIDC)
Not included
Not included
Add-on
Included
Data retentionHow long prompts and outputs are kept for abuse monitoring. Never used for training.
30 days
30 days
Zero
Custom
SOC 2 Type II report
Not included
Included
Included
Included
Audit logs
Not included
Not included
Included
Included
Region pinning
Not included
Not included
US or EU
Any region
Support
Support
Community
Email
Priority, 4h
Dedicated
Uptime SLA
Not included
Not included
99.9%
99.99%