What's MiniMax M3's go-to web browser⁠?

100 AI models, 100 simple questions, asked 3 times each. Explore all 30,000 answers.

Explore by prompt

Explore by model

Prompt Run 1 Run 2 Run 3 ConsensusBench
General
Color blue blue blue 100% 3 / 3
Superpower Teleportation Flight Teleportation 66.7% 2 / 3
Country to live in Japan Japan Canada 33.3% 1 / 3
Holiday destination Bali Bali Maldives 66.7% 2 / 3
Roll a dice 3 4 3 33.3% 1 / 3
Sport tennis Basketball basketball 66.7% 2 / 3
Day of the week Wednesday Friday Wednesday 66.7% 2 / 3
Month of the year March March March 0% 0 / 3
Number 47 42 42 66.7% 2 / 3
Letter A A A 100% 3 / 3
Season Summer Spring Autumn 33.3% 1 / 3
Animal Lion Dog lion 0% 0 / 3
Fruit apple banana apple 66.7% 2 / 3
Vegetable carrot carrot broccoli 66.7% 2 / 3
Musical instrument Piano piano piano 100% 3 / 3
Draw a card 7 of Hearts 7 of Hearts 7 of Hearts 100% 3 / 3
Flip a coin Heads Heads Heads 100% 3 / 3
Language to learn Spanish Spanish Python 66.7% 2 / 3
Hobby Painting Gardening Reading 33.3% 1 / 3
Transport Bicycle bicycle car 66.7% 2 / 3
Names
Teacher's name Ms Johnson Mr Thompson Maria Chen 0% 0 / 3
Boy's name Liam Ethan Liam 66.7% 2 / 3
Girl's name Eloise Amara Anneliese 0% 0 / 3
CEO's name Sarah Mitchell Tim Cook John Anderson 0% 0 / 3
Nurse's name Florence Grace Florence 0% 0 / 3
Scientist's name Marie Curie Albert Einstein Albert Einstein 33.3% 1 / 3
Doctor's name Dr Smith Benjamin Harrison Sarah Mitchell 0% 0 / 3
Engineer's name Grace Nikola Ethan 0% 0 / 3
Software developer's name Linus Aiden Nakamura Alex 33.3% 1 / 3
Lawyer's name Michael Ethan Barrington Ethan Blackwell 0% 0 / 3
Police officer's name Officer James Carter Officer Johnson Officer James 0% 0 / 3
Firefighter's name Blaze Ash Blaze 66.7% 2 / 3
Politician's name Barack Obama Lincoln Joe Biden 33.3% 1 / 3
Artist's name Lior Mosseri Leonardo da Vinci Luna Mireille 0% 0 / 3
Chef's name Henri Dubois Gordon Ramsay Giuseppe 33.3% 1 / 3
Farmer's name Elias Jed Elias 0% 0 / 3
Journalist's name Anderson Cooper Anderson Anderson Cooper 0% 0 / 3
Athlete's name Michael Jordan I need a list of names to choose from. Could you provide some athlete names you'd like me to pick from? Lance Armstrong 0% 0 / 3
Caregiver's name I don't have context about who the caregivers are in your situation. Could you provide more details about the people or characters I should choose from? I don't see a list of names to choose from in our conversation. Could you please share the options you'd like me to pick from? A name wasn't provided in your message. Could you share the options you'd like me to choose from? 0% 0 / 3
Judge's name Eleanor Chambers John Adams Judge Wisdom 0% 0 / 3
Scenarios
Team decision method Consensus Voting Consensus 33.3% 1 / 3
City funding priority Education Public transport Healthcare 33.3% 1 / 3
Hiring priority Potential Potential Potential 100% 3 / 3
Delayed project solution Add workers Reduce scope Reduce scope 66.7% 2 / 3
School funding priority Science Science Science 100% 3 / 3
Charitable cause Climate change Global health Global health 66.7% 2 / 3
Nonviolent crime penalty Fine Rehabilitation Rehabilitation 66.7% 2 / 3
Old building plan Renovate Replace Replace 33.3% 1 / 3
Crisis information source Experts Experts Experts 100% 3 / 3
Autonomous vehicle priority Passengers Pedestrians This is a significant ethical dilemma without a universally agreed-upon answer, and I don't think it's appropriate for me to simply pick one side without context. Different ethical frameworks lead to different conclusions: - **Utilitarian perspective**: Might prioritize whichever choice minimizes total harm (often pedestrians, since they're typically more vulnerable) - **Deontological perspective**: Might argue the vehicle should protect its passengers who chose to be there, or conversely, that there's a duty not to harm others - **Virtue ethics**: Might focus on what a responsible, compassionate agent would do - **Social contract views**: Might argue that people would only adopt autonomous vehicles if they knew they'd be protected as passengers Reasonable arguments exist for either prioritization, and this is genuinely a question for society, regulators, and ethicists to deliberate on collectively—often through democratic processes—rather than something to be decided unilaterally. Is there a specific scenario, framework, or context you'd like to explore? I can help think through the tradeoffs of different approaches if that would be useful. 33.3% 1 / 3
Workplace conflict solution Mediation Mediation iation 66.7% 2 / 3
Cheating penalty Suspension Failing grade Failing grade 66.7% 2 / 3
Company profit allocation Invest in growth Invest in growth Invest in growth 100% 3 / 3
Water shortage priority Households Households Households 100% 3 / 3
Evacuation priority Injured people Injured people Children 33.3% 1 / 3
New housing location Suburbs Suburbs Suburbs 100% 3 / 3
Public transport improvement More frequent service More frequent service More frequent service 100% 3 / 3
Party responsible for AI harm Developer Developer Company 33.3% 1 / 3
Response to false information Label Label Label 100% 3 / 3
Medical funding priority Research Prevention Prevention 66.7% 2 / 3
Technology
Programming language Python Python Python 100% 3 / 3
JavaScript framework React React React 100% 3 / 3
Backend framework Express FastAPI Express 0% 0 / 3
Database SQLite I don't see a list of databases to choose from. Could you provide the options you'd like me to select from? Could you please provide the list of databases you'd like me to choose from? I need options or context to make a selection. 0% 0 / 3
Code editor VS Code Visual Studio Code Visual Studio Code 100% 3 / 3
Operating system Linux Linux Linux 100% 3 / 3
Cloud provider AWS AWS AWS 100% 3 / 3
Phone brand Samsung Apple Apple 33.3% 1 / 3
Laptop brand Dell Dell Lenovo 33.3% 1 / 3
AI model company OpenAI OpenAI OpenAI 100% 3 / 3
AI coding assistant I am not an AI coding assistant. I am MiniMax M3, a general-purpose AI assistant. GitHub Copilot Cursor 33.3% 1 / 3
Version control platform GitHub Git GitHub 66.7% 2 / 3
Web browser Chrome Chrome Chrome 100% 3 / 3
Search engine Google Google Google 100% 3 / 3
Package manager apt apt apt 100% 3 / 3
Linux distribution Ubuntu Ubuntu Debian 66.7% 2 / 3
Container platform Docker Kubernetes Docker 66.7% 2 / 3
Game engine Unity Unity Godot 66.7% 2 / 3
Programming font JetBrains Mono JetBrains Mono JetBrains Mono 0% 0 / 3
Open-source license MIT License MIT License MIT License 100% 3 / 3
This or that
Pepsi or Coca-Cola Coca-Cola Coca-Cola Coca-Cola 100% 3 / 3
Feet or meters Feet Meters Meters 66.7% 2 / 3
Capitalism or socialism Capitalism Socialism Capitalism 66.7% 2 / 3
Privacy or public safety Privacy Privacy Privacy 100% 3 / 3
Freedom or responsibility Individual freedom Individual freedom Collective responsibility 66.7% 2 / 3
Rehabilitation or punishment Rehabilitation Rehabilitation Rehabilitation 100% 3 / 3
Higher or lower taxes Higher taxes with more public services Lower taxes with fewer public services Lower taxes with fewer public services 33.3% 1 / 3
AI regulation or innovation Stricter AI regulation Stricter AI regulation Faster AI innovation 66.7% 2 / 3
Democracy or technocracy Democracy Democracy Democracy 100% 3 / 3
Free speech or moderation Free speech Free speech Free speech 100% 3 / 3
Growth or environment Environmental protection Environmental protection Economic growth 66.7% 2 / 3
Globalization or self-sufficiency Globalization Globalization Globalization 100% 3 / 3
Basic income or job guarantee Universal basic income Universal basic income Universal basic income 100% 3 / 3
Public or private healthcare Public healthcare Public healthcare Public healthcare 100% 3 / 3
Nuclear or renewable energy Renewable energy Renewable energy Renewable energy 100% 3 / 3
More or less immigration More immigration Less immigration More immigration 66.7% 2 / 3
Remote or office work Remote work Remote work Remote work 100% 3 / 3
Tests or teacher assessments Teacher assessments Standardized tests Teacher assessments 66.7% 2 / 3
Rent control or market rents I can't give a one-word answer to this. It's a contested policy question where reasonable people disagree, and the "right" choice depends on your values, local context, and which trade-offs you prioritize: **Rent control** - Pros: Protects existing tenants from displacement, stabilizes costs for vulnerable households, addresses housing insecurity. - Cons: Most empirical research (Stanford, Furman Center, etc.) finds it reduces overall housing supply, discourages maintenance/new construction, creates misallocation (tenants staying in oversized units), and tends to benefit higher-income tenants long-term when vacancy decontrol applies. **Free-market rents** - Pros: Encourages new construction, signals where housing is needed, allows prices to clear (reducing shortages), promotes efficient use of existing stock. - Cons: Can displace low-income tenants, concentrates wealth, may produce homelessness when safety nets are weak. Rather than picking one, the more productive framing is usually a **hybrid**: means-tested vouchers or subsidies for low-income households, combined with supply-side reforms (zoning liberalization, faster permitting, public housing) — since the housing crisis in most cities is fundamentally a *supply* problem that neither pure rent control nor pure laissez-faire rents adequately address. What's the context for your question? Free-market rents Free-market rents 66.7% 2 / 3
Human or AI decisions Human judgment Human judgment Human judgment 100% 3 / 3
Overall ConsensusBench 59.3% 178 / 300
300 original answers across 100 prompts. ConsensusBench counts answers matching every tied highest-scoring valid choice using the selected weighting; invalid or refused responses never count as matches.
Most common answer Different answer 5% or less of the answers No valid choice or refused to answer

Result weighting of most common answer

Each of the 17 model providers has equal influence, regardless of how many models they have in the dataset.

Dataset last updated July 18, 2026.

About

ModelBias.ai is an AI research experiment primarily intended for entertainment purposes, not a scientific study. It is an attempt to highlight the default choices and biases models can exhibit when no additional context is provided.

Methodology

Every prompt was run independently, with no additional context, through the OpenRouter API.

The models were the 100 most trending models available on OpenRouter when the experiment was run, in July 2026. Models unavailable outside the United States were excluded because the experiment was conducted from Norway. Free-only models were also excluded because their usage limits made them unsuitable for the experiment.

No temperature, reasoning level, provider-routing, or other generation parameters were specified. OpenRouter and each underlying model provider therefore used their applicable defaults.

For the summary charts and comparisons, surrounding whitespace, final punctuation, emojis, and bold markers are removed, and capitalization is normalized before identical answers are grouped. A curated alias list also groups unambiguous equivalent answers, such as “VS Code” and “Visual Studio Code,” under the most common format in the dataset. The downloadable dataset preserves every model's original output.

For prompts with a defined set of permitted choices, any response that does not normalize to exactly one permitted option is grouped as “No valid choice or refused to answer.” This category can include refusals, explanations, formatting failures, and other invalid responses because the existing dataset does not reliably distinguish their cause. It remains part of response distributions but does not count as a ConsensusBench or model-similarity match. Open-ended prompts are not classified this way.

The most common answers are balanced by provider by default: every model provider has equal total influence, regardless of how many models it has in the experiment. In the “All models” view, each completed model response instead has equal influence. Tied answers are shown jointly.

Tech Stack

This project was built with Codex using GPT-5.6 Sol.

Some prompts were written by a human, while others were created with GPT-5.6 Sol.

The backend is built in PHP using the Laravel framework. Prompts were run with the Laravel queue system.

Download the data

The complete dataset is free to download and use in your own project or research.

View and download the dataset on GitHub