unbiased ai

LobeHub, with Pareto behind your Agents

LobeHub keeps a team’s Agents in one workspace, each with its own prompt, topics and tools. Point its OpenAI provider at Unbiased, add pareto to the model list, and any Agent can run on Pareto, which selects frontier and open models for each request and blends their outputs into one answer.

Choose Personal · $10/month See the setup →

Settings from LobeHub’s docs50M tokens a week

An illustration of LobeHub on Pareto. Not a screenshot.

Set three environment variables

Self-hosted LobeHub reads its model providers from environment variables. Unbiased speaks the OpenAI Chat Completions format, so the built-in OpenAI provider can point at it with a new proxy URL and a hand-written model list.

  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. Point the OpenAI provider at Unbiased

    On the LobeHub server, set OPENAI_API_KEY to your key and OPENAI_PROXY_URL to https://api.unbiased.ai/v1. Keep the /v1: LobeHub’s docs warn that a missing suffix can come back as empty replies.

  3. Write the model list by hand

    Set OPENAI_MODEL_LIST as shown below. -all hides OpenAI’s built-in models, +pareto=Pareto adds pareto with the display name Pareto, and <1000000:fc> sets a 1,000,000-token context and turns on tool calling.

  4. Restart and choose Pareto

    Restart LobeHub so it reads the new values, then choose Pareto as an Agent’s model. With Web Search on, the Agent reads current pages before it answers, then shows you where it looked. To search, click + below the chat input, turn on Web Search and pick Smart Search.

Settings follow LobeHub’s own documentation: LobeHub docs: Model provider environment variables.

Environment variables
OPENAI_API_KEY=your-unbiased-api-key
OPENAI_PROXY_URL=https://api.unbiased.ai/v1
OPENAI_MODEL_LIST=-all,+pareto=Pareto<1000000:fc>
Or in the app: Add Custom Provider
# App Settings › AI Providers › Add Custom Provider
Provider ID: unbiased
Provider Name: Unbiased
Request Format: OpenAI
Proxy URL: https://api.unbiased.ai/v1
API Key: your-unbiased-api-key
# Then Create Custom AI Model
Model ID: pareto
Model Display Name: Pareto
Model Type: Chat
Maximum Context Window: 1000000
Support for Tool Calling: on

Leave Use Responses API Specification and Use Client Request Mode off. On a self-hosted install, Web Search also needs a search provider on the server, SearXNG by default through SEARXNG_URL.

Your Agents stay. Their model changes.

The provider settings change. The Agents your team built, and the way they talk to people, stay where 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 LobeHub 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.

Host LobeHub yourself. Pay for the model by the month.

One API key works across LobeHub and 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.

LobeHub on Pareto, questions answered

Why doesn’t Fetch models show pareto?

Unbiased does not serve GET /v1/models, so there is no list for LobeHub to fetch. Add the model by hand: +pareto in OPENAI_MODEL_LIST, or Create Custom AI Model with Model ID pareto in the app. It is the only model string you need.

Why won’t my Agent search the web?

Check two things. Custom models start with no capabilities, so turn on tool calling for Pareto to run the search itself: :fc in the angle brackets of the model list, or Support for Tool Calling in the app. And a self-hosted LobeHub needs a server search provider, SearXNG by default, set with SEARXNG_URL. Then turn on Web Search from the + below the chat input.

Environment variables or Add Custom Provider?

Environment variables set Pareto for everyone on the server and replace the built-in OpenAI provider’s model list. Add Custom Provider adds Unbiased as its own provider, so OpenAI’s entry is left alone. Both follow LobeHub’s docs for self-hosted installs. We have not confirmed which LobeHub Cloud plans allow your own provider keys.

Can everyone on our LobeHub server share one key?

Not on a Personal plan: Personal and Personal Max are for one person. For a shared workspace, use Team, which runs on prepaid pay-as-you-go credits. Commercial or production use needs a separate written agreement.

Has this been tested inside LobeHub?

Not yet. These settings follow LobeHub’s provider and model-list documentation, and the Unbiased endpoint they call answers standard Chat Completions requests for pareto, including tool calls like the one Web Search makes. Format compatibility is not a promise that every LobeHub feature works, so tell us if one doesn’t.

Is Pareto made by the LobeHub team?

No. Pareto is Unbiased’s model. LobeHub is made by LobeHub, and Unbiased is not affiliated with or endorsed by them. You keep LobeHub 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.

Ask an Agent something it has to look up

Three variables, one key and a restart. Pareto starts at $10 a month for 50M tokens a week.

Choose Personal · $10/month See the setup →