AI inference API
Inference, turned up to eleven.
Two models, one API, honest per-token pricing. Start with free credits; commit when you're ready.
Base Eleven is a fictional company, built as a hands-on RevOps practice project. Every name, number and logo here is synthetic.
Tour
See it run
Create a key, make a first call, compare the models side by side, then watch the usage page catch up.
Models
Two models, one API
One API for fast, affordable model inference: pay as you go on a card, or commit for a discount and priority throughput.
Eleven Lite
Simulatedeleven-lite
- Input
- $2.00 per million tokens
- Output
- $10.00 per million tokens
- First token
- about 150 ms
- Speed
- 120 tokens a second
- Context
- 32,000 tokens
- Max output
- 1,024 tokens
Eleven Max
Simulatedeleven-max
- Input
- $4.00 per million tokens
- Output
- $20.00 per million tokens
- First token
- about 600 ms
- Speed
- 40 tokens a second
- Context
- 128,000 tokens
- Max output
- 4,096 tokens
A typical call, 300 tokens in and 400 out, costs $0.0046 on Eleven Lite.
Pricing
Pick how you buy
Free trial
$5 of usage at list prices, for 14 days; no card
Pay as you go
A card on file, billed monthly in arrears at list prices
Enterprise
Prepaid credits at up to 30% off list, drawn down per token
Solutions