MoreRight

The inverted LessWrong mirror. Every post, argued the other way — in inverted colors, by local models, for laughs. An idea by Utah teapot (@SkyeSharkie).

1. Get a LessWrong post

Fetch uses the LessWrong GraphQL endpoint directly from your browser. If that fails (CORS, offline, moved post), the paste box is the graceful fallback — nothing here needs a server.

2. Configure the two-pass ollama pipeline

If no ollama endpoint answers, Invert falls back to the demo on the built-in sample so you can see the whole flow. Demo output is always labelled demo output.

Which models, and why two passes

Utah teapot named the tension herself: “for better writing styles you need older models but for higher reasoning abilities you need newer models.” She is right, and that is exactly why MoreRight is a pipeline, not a prompt. Inverting an argument is a reasoning task — you have to find the load-bearing claims, negate them without strawmanning the structure, and re-read the evidence against the original conclusion. Rendering the result is a voice task — fluent, period LessWrong prose. One model is asked to do neither job; each pass gets the model class that is good at its half.

Pass 1 — Invert (newer reasoning models)

Pass 2 — Render (older, fluent, period-voice models)

Model names and tags verified against ollama.com/library on 2026-11-21: qwen3 (tags include 8b, 14b, 32b), deepseek-r1 (distills include 8b, 14b), llama3.1 (8b), mistral (7B, v0.3), llama2 (7b, 13b). Download sizes are the usual 4-bit approximations, not guarantees.

The process

  1. Fetch. Parse the post ID from the LessWrong URL and pull the post from LessWrong's GraphQL endpoint. On any failure, paste the text instead.
  2. Invert (Pass 1). The reasoning model returns an inversion map only: flipped title, flipped thesis, each load-bearing claim negated, the original's own evidence re-read against its conclusion, and the original structure preserved.
  3. Render (Pass 2). The style model rewrites the post from the map alone — same structure, opposite argument, fluent LessWrong-like voice.
  4. Label. The output is shown beside the original, in inverted colors, under the parody banner. The banner is part of the output, not decoration: it travels with the copy button.

Honest limits