Add LLM access to your platform.
Subscribe to Unlimited pools and hand out API keys, each with its own rate limits and usage tracking. We handle inference, routing, and billing.
How it works
1. Get a management key
Create an account and generate a management API key from your dashboard. This key authenticates usage tracking and other account operations.
2. Subscribe via API
Subscribe to an Unlimited pool with USDC on Base — no card required. Send the payment tx hash with your chosen pool and time blocks, and get a live API key back in the same call:
POST /api/agent/subscribe-pool
{
"txHash": "<USDC_TX_HASH>",
"poolSlug": "<pool from /api/agent/plans>",
"blocks": ["asia"]
}{
"subscriptionId": "sub_...",
"pool": "frontier",
"blocks": ["asia"],
"status": "active",
"apiKey": "sk-..."
}3. Your users call the API
Each user hits our API with their own key. Standard OpenAI SDK — just change the base URL:
from openai import OpenAI
client = OpenAI(
api_key="sk-...",
base_url="https://api.cheapestinference.com/v1"
)
response = client.chat.completions.create(
model="glm-5.2",
messages=[{"role": "user", "content": "Hello"}]
)4. Monitor usage per key
Track spend, requests, and token usage for each key via the dashboard or API:
GET /api/usage Authorization: Bearer mk_your_management_key
Who uses this
SaaS platforms — Add AI features to your product. Each of your users gets their own key with independent rate limits and usage tracking.
Agencies — Give each client their own key. Bill them separately. See usage per client.
Agent platforms — Run multiple agents, each with its own key and usage tracking. Unlimited pool access, no surprise bills.
No volume discounts. Already the lowest.
We don't do volume tiers because our prices are already at cost. Whether you run 1 key or 1,000 — same flat-rate Unlimited pool pricing. No negotiations, no sales calls, no "contact us for enterprise pricing". Each API key has a fixed concurrency limit — need more parallelism? Create more keys.