Pay per call
Send a request, get a 402 with the price, sign it, get your answer. Settled on Arc through the Arcus facilitator. No key, no account, no gas.
Arc mainnet · x402
Every model,
refracted through
one endpoint.
Point your agents at Prism402. Twenty-seven models behind an OpenAI-compatible API, metered by the token and settled in USDC on Arc. No signup, no subscription, no card.
The meter
Pay for tokens,
not for access.
No plans, no credit packs, no seats. Every call is billed for the tokens it actually used, at the price on the tag, in USDC on Arc. An agent can pay request by request without ever opening an account.
50%
off repeated calls
served from the response cache
Read the docs
served from the response cache
Prepaid keys
Deposit USDC once and get a prism_sk_ key that works in any OpenAI SDK. Every call spends exactly its token cost.
Response cache
Byte-identical deterministic calls are answered from cache at half price. Agent loops repeat themselves more than you'd think.
The change comes back
Pay-per-call signs a ceiling. What the call didn't use is booked to your wallet and paid back in batches. Close a key and the rest goes back too.
new Trench Scan
Read the contract
before the chart.
Arc's launchpads minted 83,751 tokens in a single day1. Paste any CA and Prism402 reads it straight from the chain: supply, owner, upgrade hooks, mint, blacklist and fee switches, then pool depth and 24h flow. A model explains what it found, in three sentences.
Free on the web · $0.002 per call for agents at POST /v1/scan
Recently scanned
1. BeInCrypto, "Meme Coin Launchpads Captured 82% of Arc's First Day Trading Volume", Sep 16 2026: Arguspad minted 83,751 tokens in 24 hours.
Catalog
Fresh in the spectrum
Prices in USDC per 1M tokens. What's posted is what you pay.
The token
$PRISM: hold it, pay less.
The token of the Prism402 router on Arc. Hold 1,000,000 $PRISM in the wallet that owns your key and every call is billed 20% lower, read from the chain at call time. Inference is always paid in USDC: you never need the token to use the router.
CA
Not deployed yet. The only address will appear here and on @prism402.
Drop-in
Your SDK already
speaks Prism402.
- 01
Bring a wallet, not an email
There are no accounts. The wallet that pays is the owner, and the key is the account.
- 02
Deposit USDC, mint a key
Fund a prepaid prism_sk_ key from the dashboard or any x402 client. It's shown once.
- 03
Swap one URL
Point any OpenAI client at Prism402. Or skip the key and let your agent pay each call with x402.
from openai import OpenAI client = OpenAI( base_url="https://prism402.com/v1", api_key="prism_sk_...", ) reply = client.chat.completions.create( model="kimi-k2.6", messages=[{"role": "user", "content": "What is Arc?"}], ) print(reply.choices[0].message.content)
import OpenAI from "openai"; const client = new OpenAI({ baseURL: "https://prism402.com/v1", apiKey: process.env.PRISM_KEY, }); const reply = await client.chat.completions.create({ model: "glm-5.3", messages: [{ role: "user", content: "What is Arc?" }], });
# 1. Ask. No key: the answer is 402 with the price. curl -i https://prism402.com/v1/chat/completions \ -H "content-type: application/json" \ -d '{"model":"gpt-oss-120b","max_tokens":800, "messages":[{"role":"user","content":"hi"}]}' # 2. Sign the USDC authorization in PAYMENT-REQUIRED, # resend with PAYMENT-SIGNATURE. Any x402 v2 client # (e.g. @x402/fetch) does both steps for you.
Your agents are ready. Give them the spectrum.
Start buildingBilling is exact: tokens used × posted rate, settled in USDC on Arc. No per-call fee, nothing recurring, nothing hidden.