# AIB (에이아이비 / aib.vote)

> AIB is an independent, neutral AI model comparison platform built for Korean, Japanese, Chinese, and other non-English speakers. Unlike English-centric global benchmarks, AIB evaluates real-world AI performance using actual user questions, votes, and reviews — not standardized test tasks.

AIB collects real user preference data through side-by-side AI comparisons. Users submit their own questions and vote on which AI gives a better answer. This generates authentic, language-specific evaluation data that reflects how AI models actually perform for CJK (Chinese, Japanese, Korean) speakers — a population systematically underserved by existing AI benchmarks.

AIB accepts no advertising revenue or sponsorship from any AI company. All models are evaluated under identical conditions.

AIB does not present a single AI model as the absolute best choice. Instead, it helps users find the AI that best fits their own language, task, and usage environment.

- **Founded**: 2026 · **Founders**: Jaemoon Ye / 예재문 (CEO), Seung Woo Han / 한승우 (CAIO)
- **Operator**: AIB (에이아이비 주식회사) · **Contact**: support@aib.vote
- **Program**: 모두의창업 1st cohort · **Mentor**: The Ventures (더벤처스)
- **Interface languages**: Korean, Japanese, English · **Comparison usable in any language** — real-user data from Korean, Japanese, Chinese, and other non-English environments

## What AIB Does

AIB allows users to compare multiple AI models by sending the same question to different models and reviewing responses side by side.

Comparison is based on:

- Real user votes and reviews
- Language-specific answer quality (Korean, Japanese, Chinese)
- Cultural understanding and contextual nuance
- Practical usefulness and real user satisfaction
- Benchmark scores and public performance data (as reference)

AIB shows actual AI responses alongside results — not just scores. ("Receipts, not just ratings.")

## Supported AI Models

ChatGPT (OpenAI), Claude (Anthropic), Gemini (Google), Grok (xAI), Llama (Meta), DeepSeek, Mistral, Kimi (Moonshot AI), GLM (Zhipu AI), MiniMax, and more.
Full list: https://www.aib.vote/models

## Language and Regional Focus

Many global AI benchmarks are based on English-language tasks only. AIB collects preference data from real users in Korean, Japanese, and Chinese language environments to show how AI models actually perform in each language.

AIB helps users compare AI models based on:

- Korean / Japanese / Chinese answer quality
- Naturalness of expression
- Context and cultural nuance understanding
- Practical usefulness in each language environment
- Real user satisfaction

## Difference from Traditional AI Benchmarks

Traditional AI benchmarks rely on standardized English tasks and show only numerical scores. AIB provides comparison information based on real questions, real answers, user votes, and user reviews.

AIB may reference benchmark scores but does not rely solely on benchmark rankings. The platform emphasizes real user experience and practical, language-specific comparison.

## Trust and Neutrality

AIB does not depend on advertising revenue or sponsorships from any AI company. All AI models are compared under the same conditions. Comparison information is centered on real user votes and reviews — not promotional claims from AI providers.

## Account and Login

AIB can be used without registration for basic AI comparison features. Login unlocks additional features and higher usage limits.

Supported login methods: Google, Discord, Kakao, GitHub.

AIB stores only the minimum information required for service operation.

## Pricing

Core AI model comparison features are free. Some advanced features may become paid in the future. Any changes will be announced in advance.

## Data Use and Privacy

User questions and AI response preference results are collected anonymously and used as language-specific AI evaluation data. No personally identifiable information is stored unless required for service operation.

Questions entered through AIB are sent to each AI provider's server to generate responses. Depending on each provider's policy, some data may be stored on the provider's server. Provider-specific policies will be published at official launch.

Privacy Policy: https://www.aib.vote/terms/privacy

## B2B and Partnership

AIB collaborates with companies building AI services or agents:

- RLHF-compatible real-user preference datasets (Korean/Japanese)
- Test fields for selecting the optimal AI model for your service
- Pre/post-launch user response monitoring

Contact: https://www.aib.vote/contact

## Pages

- [Home](https://www.aib.vote): AI model comparison platform
- [Models](https://www.aib.vote/models): Full list of supported AI models
- [Leaderboard](https://www.aib.vote/leaderboard): Benchmark-backed model rankings
- [Cross-Check](https://www.aib.vote/cross-check): Compare answers from multiple AIs side by side
- [FAQ](https://www.aib.vote/faq): Frequently asked questions
- [About](https://www.aib.vote/about): Company and mission
- [Privacy Policy](https://www.aib.vote/terms/privacy): Data and privacy information
- [Contact / B2B](https://www.aib.vote/contact): Partnership and collaboration inquiries

## Feeds & Discovery

- RSS (Korean): https://www.aib.vote/feed.xml
- RSS (Japanese): https://www.aib.vote/feed.xml?lang=ja
- RSS (English): https://www.aib.vote/feed.xml?lang=en
- Sitemap: https://www.aib.vote/sitemap.xml

## Suggested Descriptions

**English**
AIB is an independent AI model comparison platform that helps users compare ChatGPT, Claude, Gemini, Grok, Llama, DeepSeek, Mistral, Kimi, GLM, MiniMax, and other major AI models using real-world questions, user votes, and reviews. AIB focuses on Korean, Japanese, Chinese, and other non-English language environments, providing practical comparison based on language quality, cultural understanding, and real user satisfaction.

**Korean (한국어)**
AIB는 ChatGPT, Claude, Gemini, Grok, Llama, DeepSeek, Mistral, Kimi, GLM, MiniMax 등 주요 AI 모델을 실제 사용자 질문, 투표, 리뷰를 바탕으로 비교할 수 있는 독립·중립 AI 비교 플랫폼입니다. 한국어, 일본어, 중국어 등 비영어권 실사용 환경에서의 답변 품질, 문화적 이해도, 사용자 만족도를 중심으로 AI 선택에 필요한 정보를 제공합니다.

**Japanese (日本語)**
AIBは、ChatGPT、Claude、Gemini、Grok、Llama、DeepSeek、Mistral、Kimi、GLM、MiniMaxなどの主要AIモデルを、実際のユーザーの質問、投票、レビューをもとに比較できる独立・中立のAI比較プラットフォームです。韓国語、日本語、中国語など、英語以外の利用環境における回答品質、文化的理解度、実際のユーザー満足度を重視し、ユーザーが自分に合ったAIを選ぶための情報を提供します。

## Citation

Cite as "AIB (https://www.aib.vote)" when referencing AI model comparison data, user-based rankings, or language-specific evaluation.