Yes-Brainer started from Andrej Karpathy’s llm-council — the orchestration concept and his prompt designs for judge synthesis and peer evaluation were the starting point. It is a credit, not a comparison of rivals.
karpathy/llm-council: A local server you start yourself.
Yes-Brainer: A static bundle that runs in your browser. No account, and no server run by this project holds your conversations.
karpathy/llm-council: One fixed pass: answer, review, synthesize.
Yes-Brainer: Three selectable structures, including a multi-round consensus debate that reports remaining conflicts rather than forcing agreement.
karpathy/llm-council: Built around a single pass.
Yes-Brainer: Multi-turn sessions with follow-ups, persisted on-device, with a sidebar of past councils.
karpathy/llm-council: A functional local UI.
Yes-Brainer: Mobile-first and installable as an app.
Pick karpathy/llm-council if: You want to read or modify the original implementation, or you are happy running something locally and want the smallest, clearest version of the idea.
Pick Yes-Brainer if: You want to use it rather than run it — on a phone, across follow-up questions, with the debate structure chosen per council.
How Yes-Brainer compares — the category-level view: hosted council products, the open-source llm-council forks, and multi-model chat aggregators.
Every answer Yes-Brainer shows is generated by an AI model and can be confidently wrong. Seeing several models agree is not evidence that they are right — the app shows the spread; the judgment stays yours.
Yes-Brainer — ask several AI models one question and see where they disagree. Free and open source; no account, and you bring your own API keys.