Gondolabeta
⌁ THE VENICE INFERENCE MARKETPLACE

The same Venice models, up to 65% cheaper.

And idle DIEM, earning ~13% APR in USDC at full utilization.

Gondola is the OpenAI-compatible gateway to Venice AI. Buyers get the full Venice catalog for less, and DIEM holders turn idle daily allowance into USDC yield.

Lender APR is annualized from recent settled routing at full utilization, after the 2% platform fee. Actual earnings depend on buyer demand, your pricing, available allowance, and routing.

Open chatStart lending
OpenAI-compatibleUSDC on Base2% buyer / 2% lenderTop users pay as little as 0.3% (from 2%)Gondola never holds your DIEM
WHAT IS GONDOLA

Gondola is a marketplace for Venice AI inference. Staked DIEM on Venice mints a daily inference allowance whether its holder spends it or not, so DIEM holders deposit a Venice inference key, set one price factor against Venice's live list price, and let Gondola sell that idle allowance as paid inference for USDC yield. Buyers call a single OpenAI-compatible gateway and pay per request in USDC on Base, while the broker meters each call and routes it to the cheapest healthy key for the requested model, so the same Venice models can cost less than going direct.

HOW IT WORKS

Two sides, one gateway.

FOR LENDERS
Idle DIEM allowance, turned into USDC yield.
1
Deposit a Venice key
Connect your Base wallet and add a Venice INFERENCE API key, or mint one against your staked DIEM.
2
Set your price
Choose a single price factor against Venice's list price. Cheaper offers win routing first.
3
Earn USDC
Gondola routes paid inference to your key and accrues USDC for every request, withdrawable anytime. It never takes custody of your DIEM: your stake stays on Venice, and only the idle daily allowance is sold.
FOR BUYERS
Cheaper Venice inference, paid per request in USDC.
1
Open the chat or mint a key
Use the in-browser chat with no setup, or mint a Gondola API key for curl, the OpenAI SDK, or your editor.
2
Top up USDC on Base
Add USDC to your Gondola balance once. The same balance powers both the chat and your API keys.
3
Pay per request
Each call debits the metered cost in USDC. The broker routes to the cheapest healthy lender for your model, so you pay less than going direct to Venice.
DROP-IN ENDPOINT

Same shape as OpenAI

Point any client at api.gondola-ai.com/v1 with a Gondola key. Chat completions, image generation, video generation, and the models list are live today.

from openai import OpenAI

client = OpenAI(
  base_url="https://api.gondola-ai.com/v1",
  api_key="gnd_live_...",
)

resp = client.chat.completions.create(
  model="llama-3.3-70b",
  messages=[{"role": "user", "content": "Ciao"}],
)
print(resp.choices[0].message.content)
PRICING

A symmetric 2% fee, nothing else

Buyers pay the posted price plus 2%, lenders receive the posted price minus 2%, and the broker keeps the spread. Lenders price against Venice's live list, and because the broker always routes to the cheapest healthy key, competition pushes prices below going direct to Venice. No subscriptions, no minimums.

Heavy users pay much less: weekly placements drop your side of the fee from 2% to as little as 0.3%. How points work →

See model prices →Browse the marketplace →Explore characters →

Step aboard.

Buy Venice inference for less, or put your idle allowance to work.

Open chatStart lending