ᚠ

Start free.
Scale infinitely.

No credit card required. No wallet. Just an API key to the decentralized network.

ᚱᚢτᚾᛊ
ᚨ
Free
$0
No deposit needed
  • 1,000 requests/day
  • Inference only (all 532+ models)
  • OpenAI-compatible
  • No credit card or wallet required
  • Community support
Get Free Key →
ᛞ
Pay As You Go
~$0.0001/req
Deposit USDC on Base
  • All endpoints: inference, search, verification
  • Varies by model complexity
  • No monthly commitment
  • No minimum deposit
  • Top up anytime
Deposit USDC →
ᛊ
Pro Credits
$50+ deposit
Same per-request pricing, more features
  • All endpoints + fine-tuning + pipelines
  • 100 GB decentralized storage
  • Priority routing, lower latency
  • Cross-subnet pipelines
  • Usage analytics dashboard
Unlock Pro →
ᛃ
Enterprise
Custom
Contact us
  • Confidential compute (Intel TEE)
  • SLA + dedicated support
  • Volume discounts
  • Custom model hosting
  • SSO + team management
  • Unlimited storage
Contact us →
ᛟᚨτᛃᚾ

Frequently asked

How does billing work?
Deposit USDC on Base. Every API call deducts from your balance. No subscriptions, no monthly bills.
What if my credits run out?
API returns 402. Top up anytime — credits appear in ~30 seconds.
Can I use it for free?
Yes. 1,000 requests/day, no deposit needed. Just sign up and get an API key.
How does confidential compute work?
Add the header X-Confidential: true to any inference request. Your prompt is routed through Intel TDX Trusted Execution Environments on SN4 Targon. Nobody — not miners, not us — can see your data. Enterprise tier only.
What is a cross-subnet pipeline?
A pipeline chains multiple Bittensor subnets in a single API call. For example: collect data (SN13) → fine-tune a model (SN56) → deploy it (SN64). Available on Pro Credits and Enterprise tiers.