unbiased ai

Chatbox, answering with Pareto

Chatbox is a chat window that goes where you do: Windows, Mac, Linux, iPhone, Android and the web. Add Unbiased as a custom provider, type pareto as the model, and any chat set to it runs on Pareto, which selects frontier and open models for each request and blends their outputs into one answer. Paste in the email thread you have been avoiding and ask for the reply.

Choose Personal · $10/month See the setup →

Settings from Chatbox’s docs50M tokens a week

An illustration of Chatbox on Pareto. Not a screenshot.

Add Unbiased as a custom provider

Chatbox speaks the OpenAI Chat Completions format to any provider you add. Unbiased accepts that format, so Pareto goes in as a custom provider with its own host, key and a model ID you type yourself.

  1. Get an Unbiased API key

    Choose a plan, then create a key in the Unbiased platform. New accounts are reviewed and approved by a person before the key works.

  2. Add the provider

    In Chatbox, open Settings from the sidebar, go to Model Provider and click Add. Name it Unbiased and set API Mode to OpenAI API Compatible.

  3. Fill in the host and key

    Paste your key into API Key and set API Host to https://api.unbiased.ai/v1. Leave API Path empty: Chatbox adds /chat/completions itself.

  4. Type the model by hand

    Under the provider’s models, click New and enter pareto as the Model ID. Skip Fetch: Unbiased does not serve a model list for it to read.

  5. Save and check

    Save, press Check, then start a new chat and choose pareto under Unbiased in the model picker.

Settings follow Chatbox’s own documentation: Chatbox docs: model provider setup.

Settings › Model Provider › Add
Name:      Unbiased
API Mode:  OpenAI API Compatible
API Key:   your-unbiased-api-key
API Host:  https://api.unbiased.ai/v1
API Path:  (leave empty)
# Models › New
Model ID:  pareto
Optional model fields
# Leave blank, or fill in Pareto's limits
Context Window:     1000000
Max Output Tokens:  131072

A model added without capability boxes ticked is treated as text only, which is all a pasted email thread needs. To attach images or let the model use tools, tick Vision or Tool use on the pareto entry. Check also tests images and tools, and ticks those boxes itself if they pass.

Your chats stay. The model behind them changes.

One new provider entry in Settings. Your conversations, prompts and the apps you use them in carry on as before.

One model string, several models behind it

Pareto is Unbiased’s blended AI model. On each request it selects frontier and open models and blends their outputs into one answer, so Chatbox talks to one model string, pareto, and never has to pick.

“When the frontier improves, Pareto improves; when frontier prices fall, you should benefit.” From the pricing page.

How a Pareto answer gets made →
An illustration, not a wiring diagram. The blend includes frontier and open models, and its composition can change.

What a hard task costs on Pareto

The model card publishes Pareto’s price per finished task on two public agent benchmarks, alongside the price per token.

Terminal-Bench 4.0Tasks an agent completes in a terminal
$0.48/task
50.8% solved
DeepSWE v1.1Changes an agent makes in a real codebase
$0.24/task
69.9% solved

Pareto 26.10, preliminary October 1, 2026 runs that may change; denominators and the cost method still require confirmation. Cost is the reported mean USD per task at API rates, not an estimate for your workload. Model card and method.

Keep Chatbox. Pay for the model by the month.

One API key works in Chatbox on every device and in any other tool you point at Pareto. Pick a monthly allowance, or pay per token on Team.

Personal

$10/month

  • 50M tokens a week
  • 20 requests per minute
  • For one person
Choose Personal

Personal Max

$100/month

  • 600M tokens a week
  • 60 requests per minute
  • For one person
Choose Personal Max

Team

Pay as you go

  • Prepaid credits at published rates
  • 300 requests per minute
  • No weekly token cap
Get started with Team

Subscriptions are for one person and are not unlimited. New API accounts are reviewed by a person before access turns on, and commercial or production use needs a separate written agreement. Rates and limits are on Pricing.

Chatbox on Pareto, questions answered

Why doesn’t Fetch find pareto?

Fetch asks the provider for its model list, and Unbiased does not serve GET /v1/models, so there is nothing to fetch. Click New instead and type pareto as the Model ID. It is the only model string you need.

Which API Mode should I choose?

OpenAI API Compatible, with https://api.unbiased.ai/v1 as the host and an empty path. That is the Chat Completions format Chatbox’s guide describes for custom providers. Unbiased also serves Anthropic Messages and OpenAI Responses requests, but we have not tried Chatbox’s other modes against it.

Is this the same as Chatbox AI?

No. Chatbox AI is Chatbox’s own paid service with its own license. This setup uses Chatbox with your own key, so requests for pareto go to Unbiased and your Unbiased plan pays for them. Chatbox AI and any other providers stay available in the model picker.

Can I use the same key on my computer and my phone?

Yes, as long as they are all yours. Add the same provider, with the same host, key and Model ID, in each Chatbox app you use. Personal and Personal Max are for one person, so the key should not be shared with anyone else.

Has this been tested inside Chatbox?

Not yet. These settings follow Chatbox’s provider guide, and the Unbiased endpoint they call answers streaming Chat Completions requests for pareto. Format compatibility is not a promise that every Chatbox feature works, so tell us if one doesn’t. If you are comparing chat apps, see Pareto as a ChatGPT alternative.

Is Pareto made by the Chatbox team?

No. Pareto is Unbiased’s model. Chatbox is made by the Chatbox team, and Unbiased is not affiliated with or endorsed by them. You keep Chatbox and point it at the Unbiased API.

What happens when I hit the weekly allowance?

Requests over your plan’s limits return an HTTP 429 error. Personal includes 50M tokens a week and 20 requests per minute. Personal Max raises that to 600M a week and 60 requests per minute for $100/month. Team uses prepaid pay-as-you-go credits with no weekly token cap.

Do I choose between models?

No. You send every request to one model string, pareto. Pareto selects frontier and open models for each request and blends their outputs into one answer.

Is my code kept?

Personal and Personal Max are not our zero data retention tier: prompts and responses may be kept up to 30 days, as section 5(c) of the Terms explains. For work that must not be kept, ask about the Zero Data Retention API on the Private AI page.

Pareto works in other tools too

The same key and the same model string go into each of these. See them all.

Product names belong to their owners. Except where a guide says otherwise, Unbiased is not affiliated with or endorsed by the makers of the tools on this page.

Answer the email you’ve left on read

One provider, one key and the model ID typed in by hand. Pareto starts at $10 a month for 50M tokens a week.

Choose Personal · $10/month See the setup →