Plans

Get 10% off when you pay annually.

MonthlyAnnually

Free

$0
Test out the Arli platform
  • Unlimited Tokens!
  • Unlimited Requests!
  • 1 request at a time

  • Access Qwen3-14B Models

  • 5 times / 2 days trial of all models

  • Delayed Response

  • Slower with more requests

  • No Image Generation Tools

Select

Starter

$10/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 1 request at a time

  • Access Up to 32B Models

  • No Image Generation Tools

Select

Core

$15/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 1 request at a time

  • Access Up to 235B Models

  • Image Generation Tools

Select

Advanced

$25/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 2 parallel requests

  • Priority Response

  • Access Up to 235B Models

  • Image Generation Tools

Select

Professional

$75/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 6 parallel requests

  • Priority Response

  • Access Up to 235B Models

  • Image Generation Tools

Select

Ultimate

$250/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 20 parallel requests

  • Priority Response

  • Access Up to 235B Models

  • Image Generation Tools

Select

Continuous

$500/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 20 parallel requests

  • Highest Priority

  • No load balancing slowdowns

  • Ideal for continuous usage

  • Access Up to 235B Models

  • Image Generation Tools

Select

Continuous2

$1000/month
(USD rough conversion from IDR)
  • Unlimited Tokens!
  • Unlimited Requests!
  • 40 parallel requests

  • Highest Priority

  • No load balancing slowdowns

  • Ideal for continuous usage

  • Access Up to 235B Models

  • Image Generation Tools

Select

Enterprise

Contact us for dedicated deployments

Lowest cost of any inference provider
  • Dedicated GPUs
  • No rate limiters
  • Unlimited parallel requests

  • Reliability increase

  • Fast Response Speed

  • Custom Model or LORA

Contact Us