ArcaArca
Go to App →

The on-chain inference gateway
for AI agents.

Point your agents at one endpoint: one key to the frontier model catalog, routed and cached to cut token cost by up to 30%, metered by the token and settled in USDG on Robinhood.

Under the hood

—

Tokens served

—

Inference calls

—

Models online

—

Uptime

The gateway

One gateway.
Cheaper tokens.

Arca is an inference gateway: one endpoint and one key in front of the frontier model catalog. We don't pretend to own the GPUs, we route to the providers that do, and the mechanisms on the right are how the same call comes back cheaper than going direct.

up to30%lower token cost
via routing + response cache
Read the whitepaper →
A

Smart routing

Every call routes to the best live capacity serving that model. When a cheaper or faster route exists, your traffic takes it, no code change on your side.

B

Response cache

Identical repeat calls, retries, evals, loops, come back in ~10ms at half-price tokens. A posted billing rule, not a trick.

C

Tier routing

The catalog is tiered High / Balanced / Eco, so agents buy exactly the intelligence a task needs. The cheapest token is the frontier token you didn't need.

D

No platform overhead

No subscription, no per-call fee, no credit packs. You pay tokens times the posted rate, and the posted rate is the whole price.

Fresh in the catalog

The latest models routed through the gateway, every one at posted per-token rates.

Browse all models →

The token

Hold $ARCA.
The network pays you.

Every call the network serves earns revenue on-chain. Half of it is split across $ARCA holders every month, pro-rata to the supply you hold.

50%

of every call’s revenue fills the holder pool

Monthly

pro-rata payouts, straight to holding wallets

0

staking, lockups, or claim races

How rewards work →About $ARCA

Drop-in

Your SDK already
speaks Arca.

1

One email, one wallet

Sign in and a non-custodial wallet spins up in the background, no seed-phrase ceremony.

2

Load USDG, mint a key

Top up once and mint your ar_sk_ key. It's shown a single time, after that, it's yours to guard.

3

Swap one URL

Keep the OpenAI client you already ship. Point base_url at Arca and every call meters and settles itself.

python · drop-in OpenAI client
from openai import OpenAI

client = OpenAI(
    base_url="https://api.arcagateway.com/v1",
    api_key="ar_sk_live_…",
)

resp = client.chat.completions.create(
    model="glm-5.2",
    messages=[{"role": "user", "content": "Hello"}],
)
# metered + settled on-chain automatically

Your agents are ready. Give them a gateway.

Billing is exact: tokens used × posted rate, settled in USDG on Robinhood, no per-call fees, no gas, nothing recurring, nothing hidden. High-tier models always run pay-as-you-go.