# Use Pareto in Venice: instructions for the agent

You are an AI agent helping someone use Pareto, UnbiasedAI's model, in Venice. Venice is made by Venice.ai, which lists Pareto among its models and bills its own users. Follow these steps in order and stop at any mismatch.

- Human page: https://unbiased.ai/works-with/venice/
- Status: Untested by UnbiasedAI. Steps follow Venice's own screens and https://venice.ai/models/pareto, checked November 2026. Tell the user this before you start, and treat the verify step as required.
- Mode: Guided, inside Venice: the user picks Pareto in Venice's own screens. You read them each step and wait. Nothing is installed and no file changes.
- Getting a key: not needed. The user's account and bill are with Venice.ai. Pareto in Venice does not use `UNBIASED_API_KEY`.

## Quick answer

To use Pareto in Venice, open venice.ai/chat?model=pareto, choose Continue on the Pareto card, log in to Venice and send a message. Venice bills you through your Venice account, and you do not need an UnbiasedAI account.

- Name in Venice's model list: Pareto
- Account and billing: Your Venice account
- UnbiasedAI account: Not needed
- Context window: 1M tokens
- Venice privacy tier: Anonymized

## What Pareto is

Pareto is Unbiased’s blended AI model: it selects frontier and open models for each request and blends their outputs into one answer. In Venice it is one name in the model list. Venice.ai sets the price and bills the user; UnbiasedAI's own plans and rates (https://unbiased.ai/pricing/) do not apply inside Venice.

## Safety rules

- Never ask the user to paste a password, card number, sign-in code or API key into chat, and never print, echo or log one.
- The user signs in to Venice and pays Venice.ai themselves, in their own browser or app. Only they choose a plan or buy credits.
- Do not ask for, look for or use `UNBIASED_API_KEY`. It is not part of this.
- This changes no files. If a step seems to need one, stop and ask.
- If something does not match what this runbook says, stop and report it. Do not invent workarounds.
- You do not click or type in the user's apps unless they ask and your tools allow it. Read each step, then wait.

## 1. Ask what the user already has

Ask whether the user has a Venice account, and whether it has a paid plan or credits. They do not need an UnbiasedAI account. If they have no Venice account, the steps below create one.

## 2. Walk through the steps

Venice holds your account and bills you. UnbiasedAI makes Pareto and serves it to Venice. There is no key to paste and nothing to install.

You cannot click for the user. Read them one step at a time and wait for "done" after each.

1. **Open Pareto in Venice.** Go to venice.ai/chat?model=pareto. Venice opens Classic Chat and shows its card for Pareto. Choose Continue.
2. **Log in or create a Venice account.** Above the message box, Venice asks you to log in or create an account before you can use Pareto. The account is with Venice. UnbiasedAI has no part in it.
3. **Check your Venice credits.** Venice's Pareto page says Pareto is billed per token, with no subscription required, and lists the current price. It also says new Venice accounts include a free daily allowance and welcome credits, with no credit card required. Venice's paid plans each include monthly credits.
4. **Send a first message.** Check that the model menu under the message box shows Pareto. If it shows another model, open the menu and search for Pareto. Then ask for something small, such as a three-bullet summary of a document. You choose Pareto once, and it picks the models behind each answer.

Note from the human page: The Pareto on Venice launch post covers who bills you and where your prompt goes. Venice runs the app, your account and your bill, so questions about plans and credits go to Venice. Questions about Pareto itself go to info@unbiased.ai.

## 3. Verify with one real message

Ask the user to send this message in Venice with Pareto chosen: "Reply with the single word ok". Ask them what came back and which model name Venice shows on the reply. If Pareto is not in the model list, or Venice asks for something this runbook does not mention, stop and report it.

You are done only when a real reply has come back with Pareto chosen.

## Optional: Venice's API

Only if the user asks to call Pareto from code. The key is Venice's, never an UnbiasedAI key; the user sets it in their own shell.

**Or call it through Venice's API (shell)**

```sh
curl -sS https://api.venice.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $VENICE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"pareto","messages":[{"role":"user","content":"Reply with the single word ok"}]}'
```

## Where the prompt goes

A message you send to Pareto in Venice passes through two companies and then the models behind Pareto. This is what each one does with it.

- **Venice.** Venice holds your account and takes your payment. It says your chats are stored only on your device. Venice's Pareto page says it runs Pareto under its anonymized privacy tier, where prompts are not stored, profiled or tied to a personal history. Venice's privacy page has its full terms.
- **UnbiasedAI.** Your prompt reaches UnbiasedAI under Venice's own account. UnbiasedAI has no account for you and never bills you. Venice's traffic is designated ZDR, zero data retention: UnbiasedAI doesn't retain, access or train on prompts or responses after inference.
- **The models behind Pareto.** Pareto selects frontier and open models for the request and blends their outputs into one answer. Those model providers process the prompt under UnbiasedAI's own arrangements with them. The Private AI page sets out who the ZDR commitment binds.

## Roll back

- You changed no files. To stop using Pareto, the user picks another model in Venice.
- Plans and credits are managed in the user's Venice account.

## Troubleshooting

- **Do I need an UnbiasedAI account?** No. You choose Pareto inside Venice with your Venice account. An UnbiasedAI account is only for calling Pareto from your own tools or code.
- **What does Pareto cost in Venice?** Venice sets the price. Its Pareto page says Pareto is billed per token, with no subscription required, and lists the current rates. UnbiasedAI's own plans and rates don't apply inside Venice, and UnbiasedAI never bills you for it.
- **I can't find Pareto in the model list. Where is it?** The link in step 1 opens Venice's Classic Chat with Pareto already chosen. To find it yourself, open the model menu under the message box in Classic Chat and search for Pareto. If it isn't there, contact Venice, which runs the model list.
- **What does UnbiasedAI keep from my chats?** No prompt or response content after the reply is produced. Venice's traffic is designated ZDR, so UnbiasedAI doesn't retain, access or train on prompts or responses after inference. The Private AI page covers ZDR and the model providers behind Pareto.
- **Can I call Pareto through Venice's API?** Yes. Venice's API uses the OpenAI format. Create an API key in your Venice account, send requests to `https://api.venice.ai/api/v1` and set the model to `pareto`. Venice bills the usage.
- **Who do I contact if something goes wrong?** For your account, plan, credits or the app, contact Venice. For Pareto's answers, email info@unbiased.ai.
- **Is Pareto made by Venice.ai?** No. UnbiasedAI makes Pareto. Venice.ai builds and runs Venice, lists Pareto among its models and bills its own users. UnbiasedAI serves the model to Venice.

## Report back

Tell the user: the reply that came back in the verify step and anything you skipped. Account, plan and credit questions go to Venice.ai. Questions about Pareto itself go to info@unbiased.ai, with the failing step and the error text, never a password or key.
