A platform for LuciditySH models
An OpenAI-compatible API for the Synth language model family, with automatic fallback between models built in.
One request
Sign up, create a key on the API keys page, and this works:
curl https://platform.lucidity.sh/v1/chat/completions \ -H "Authorization: Bearer sh-live-••••••••••••7f2a" \ -H "Content-Type: application/json" \ -d '{ "model": "synth-2.5-pro", "messages": [{"role": "user", "content": "Hello!"}] }'
Models
Language models are billed per million tokens. Requests for synth-2.5-pro automatically fall back to synth-2.5-flash if it's temporarily unavailable. See "Automatic fallback" in the docs for details.
| Model | Slug | Type | Input | Output |
|---|---|---|---|---|
| Synth 2.5 | synth-2.5-pro | Language | $0.10 / 1M | $0.20 / 1M |
| Synth 2.5 Flash Preview | synth-2.5-flash | Language | $0.03 / 1M | $0.10 / 1M |
How billing works
Pay as you go
There is no plan to pick and no monthly statement. You buy credit, requests draw it down at the rates above, and the balance stays there until you spend it.
Metered per token
Input and output tokens are priced separately. Token counts come from the upstream model where it reports them, and are estimated from the response length where it doesn't.
Top up when you need to
You buy credit through PayPal from the Top up page. Usage and your remaining balance are both visible in the dashboard, so you can see what a run cost before you buy more.
Start with a free account
Create one, make a key, and read the docs from inside the dashboard.