Standard requests
chat/completions, streaming, curl / Python / Node. Set base URL and key. No custom SDK.
Product
One key, one model id. Call chat completions from your server. Capacity grows underneath without changing your integration.
chat/completions, streaming, curl / Python / Node. Set base URL and key. No custom SDK.
$297/mo. Same bill every month. No token invoice, no surprise overage line. Vision is included; speech and transcription are a +$100/mo add-on.
Send photos and document scans with chat. Add voice for speech and transcription through the audio endpoints, on the same key.
Access
Stripe checkout. No credentials until payment clears.
Base URL, model, limits, and key in one place. Member guide included.
Use model legia-max. Keep the key on the server, never in a public frontend.
Request shape
Placeholder only. Real base URL and key appear in the dashboard after subscribe.
curl YOUR_BASE_URL/v1/chat/completions \
-H "Authorization: Bearer YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "legia-max",
"messages": [
{"role": "user", "content": "Hello"}
]
}'Add-on
Spoken replies for +$100/mo. Enable in the dashboard. Same account and base URL as chat.
Pricing
Limited seats on shared capacity. Built for product traffic, not bulk scrapers.
Standard
Flat rate for chat and vision. Personal key, usage counters, and the full member guide.
Stripe checkout. Cancel in the customer portal anytime.
legia-max · streamingFAQ
Only in your dashboard after subscribe. Marketing pages never show production endpoints.
Any HTTP client that can call chat completions: curl, Python, Node, and most SDKs that accept a custom base URL.
One monthly price for chat and vision. Speech and transcription are a +$100/mo add-on. Rate limits protect shared capacity so every seat stays usable.
Yes. Send an image in the same chat call, the way any OpenAI-compatible client does. Vision is included. The +$100/mo voice add-on includes speech and audio transcription through the standard audio endpoints.
No. The model id and your key stay put when we add hardware. You keep shipping product; we run capacity.
Keep the key on your server. Cap max tokens. Handle 401, 429, and 5xx with backoff.