# karpathy/llm-council

LLM Council works together to answer your hardest questions

Repository: https://github.com/karpathy/llm-council
Canonical: https://ross.abutalabs.com/products/llm-council
Language: Python
License Family: other
Last push: 2025-11-22T23:35:21+00:00

## Health v2 (maintenance only)
Score: 40/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 53, release rhythm 35, longevity 20
- inputs: {"age_days": 284, "days_push": 284, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 24369, forks 4300 (observed 2026-08-28T04:11:37.040838+00:00)

## What it is
A local web app that sends a user's question to multiple LLMs via OpenRouter, has them anonymously review and rank each other's answers, and uses a Chairman model to synthesize a final response. It was built as a fun weekend hack by Andrej Karpathy and is explicitly provided as-is without support.

## Use cases
- compare answers from multiple LLMs side by side
- get a second opinion on hard questions from several AI models
- have LLMs review and rank each other's responses
- run a local ChatGPT-like interface backed by multiple providers
- evaluate which LLM gives the most accurate and insightful answers
- aggregate multiple model opinions into one final answer

## When to choose
- you want to query several frontier LLMs at once and compare their answers
- you want anonymized cross-review between models to reduce single-model bias
- you want a self-hosted, hackable ChatGPT-style UI you can modify with an LLM
- you already have an OpenRouter API key and credits

## When to avoid
- you need a supported, maintained product with bug fixes and features
- you want a single fast/cheap LLM response rather than multi-model fan-out
- you can't use OpenRouter or don't want to pay per-query API costs
- you need production-grade security, auth, or deployment support

## Facets
- artifact type: application
- maturity: maintenance
- function: llm-inference, chat-interface, web-framework, http-client
- domain: large-language-models, artificial-intelligence, web-development, developer-tools
- platform: python, self-hosted, cross-platform
- tags: openrouter, multi-llm, llm-comparison, peer-review, ensemble, chatgpt-alternative, fastapi, react, nodejs, web-server

## Member repositories
- karpathy/llm-council (main) score 40

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:37.040838+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:56:15.022223+00:00, confidence not recorded.
  - readme: https://github.com/karpathy/llm-council (fetched 2026-08-28T04:11:37.040838+00:00, sha 6c1d4af8f81d)
- Data as of 2026-08-30T08:39:29.467469+00:00.