unbiased ai

Pareto in the GitHub Copilot model picker

Copilot Chat in VS Code lets you bring your own model, and its Custom Endpoint provider accepts any URL that speaks Chat Completions. Add one entry to chatLanguageModels.json and Pareto takes a seat in the model picker. For each request, it selects frontier and open models and blends their outputs into one answer.

Choose Personal · $10/month See the setup →

Settings from the VS Code docs50M tokens a week

An illustration of GitHub Copilot in VS Code on Pareto. Not a screenshot.

Add Pareto as a Custom Endpoint model

Copilot Chat keeps bring-your-own-key models in chatLanguageModels.json. The Custom Endpoint prompts start the entry for you; you add the model id, the full URL and a line about tool calling.

  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. Open the Language Models editor

    In the Chat view, open the model picker and select Manage Language Models (the gear icon), or run Chat: Manage Language Models from the Command Palette. Select Add Models, then Custom Endpoint. It replaces the older OpenAI Compatible provider, which is deprecated.

  3. Answer the prompts

    Enter Unbiased as the group name, then a display name and your Unbiased API key. For the API type, choose Chat Completions.

  4. Fill in the model

    VS Code opens chatLanguageModels.json. Make the entry match the one on this page and save. Unbiased does not publish a model list, so leave the provider-level url out and name pareto by hand in models. Keep toolCalling set to true, or the model won’t appear in the picker for agent mode.

  5. Pick Pareto in chat

    Open the model picker and choose Pareto under Unbiased. If it isn’t there yet, restart VS Code, as the docs suggest.

Settings follow GitHub Copilot in VS Code’s own documentation: VS Code docs: Language models, Add a custom endpoint model.

chatLanguageModels.json
[
  {
    "name": "Unbiased",
    "vendor": "customendpoint",
    "apiKey": "${input:unbiasedApiKey}",
    "apiType": "chat-completions",
    "models": [
      {
        "id": "pareto",
        "name": "Pareto",
        "url": "https://api.unbiased.ai/v1/chat/completions",
        "toolCalling": true,
        "vision": false,
        "contextWindow": 1000000,
        "maxOutputTokens": 131072
      }
    ]
  }
]
Or use the Messages API type
// Change these two lines in the entry above
"apiType": "messages",
"url": "https://api.unbiased.ai/v1/messages"

The apiKey line refers to the key you entered in step 3 through an input variable, so the key itself stays out of the file. Write the full URL, path included: without a path, VS Code adds one for you.

Copilot stays. The chat model changes.

Pareto answers in chat and agent mode. The parts of Copilot that run on GitHub’s own service carry on as they were.

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 GitHub Copilot in VS Code 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 an agent task costs on Pareto

Agents make many model calls per task, so the price of a finished task matters more than the price of a token. The model card publishes both for Pareto on two public agent benchmarks.

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.

Pay for Copilot Chat on Pareto by the month

One API key works in VS Code 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.

GitHub Copilot in VS Code on Pareto, questions answered

Do I need a GitHub Copilot plan?

Not for chat. The VS Code docs say bring-your-own-key models work without a Copilot plan and without signing in to GitHub. Inline suggestions, semantic search and features that rely on embeddings still need Copilot. If you aren’t signed in, the built-in utility models aren’t available, so set chat.utilityModel and chat.utilitySmallModel to a model you added (for titles, summaries and commit messages).

Why not let VS Code find the model by itself?

If the provider entry has a url, VS Code asks that endpoint for its list of models. The Unbiased API does not implement that request yet (it returns 404), so model discovery fails. Leave that field out, put pareto in the models array and VS Code uses it as written.

Should I choose Chat Completions, Responses or Messages?

Chat Completions is the one this page sets up. Messages should also work against Unbiased’s Anthropic-compatible endpoint; change the apiType and url as shown. Responses is served at https://api.unbiased.ai/v1/responses too, but this page has not checked it with Copilot. We have not run Copilot Chat on Pareto end to end ourselves, and format compatibility is not a promise that every feature works, so tell us if one doesn’t.

My company pays for Copilot Business. Can I still do this?

Only if your administrator allows it. On Copilot Business and Enterprise, the Bring Your Own Language Model Key in VS Code policy on GitHub.com must be enabled before you can add models. Individual plans, and people with no plan, can add them directly.

How do I switch back?

Pick another model in the chat picker. To remove Pareto entirely, delete the Unbiased entry in the Language Models editor or in chatLanguageModels.json.

Is Pareto made by the GitHub Copilot in VS Code team?

No. Pareto is Unbiased’s model. GitHub Copilot in VS Code is made by Microsoft and GitHub, and Unbiased is not affiliated with or endorsed by them. You keep GitHub Copilot in VS Code 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.

Put Pareto in the picker and hand it a failing test

One JSON entry, one key and the editor you already have open. Pareto starts at $10 a month for 50M tokens a week.

Choose Personal · $10/month See the setup →