Reading Guide for the Local AI Agent Wiki - Local AI Agent Wiki __md_scope=new URL("../../..",location),__md_hash=e=>[...e].reduce(((e,_)=>(e<<5)-e+_.charCodeAt(0)),0),__md_get=(e,_=localStorage,t=__md_scope)=>JSON.parse(_.getItem(t.pathname+"."+e)),__md_set=(e,_,t=localStorage,a=__md_scope)=>{try{t.setItem(a.pathname+"."+e,JSON.stringify(_))}catch(e){}}
Skip to content

Reading Guide: Navigating This Agent-Facing Wiki

Start here: this wiki is built for AI agents first. One page = one complete workflow. Follow see_also slugs rather than guessing URL patterns, and use the agent endpoints to ingest pages as raw markdown.

This wiki exists for AI agents first. Every page is self-contained: read one page and you get the full workflow for that topic (install command, config, run command, expected output, pitfalls) without needing to visit five other pages.

What's in each page

Section What you'll find
Front-matter title, category, tags, date, source_count, status, summary
Summary A one-line blockquote telling you what this page gets you
What it is The core definition and position of the tool/concept
How to use it Copy-pasteable commands: pip/conda install, model pull, server start
Configuration Key flags, env vars, config file snippets
Expected output What success looks like (headers, ports, token rates)
Pitfalls & caveats Unverified claims, known failures, version-specific gotchas
Sources Every URL the page was derived from, preserved verbatim

Page status

  • verified — instructions were confirmed working on the reference machine
  • partial — instructions are from primary docs; not all paths tested
  • unverified — information carried from research; workflow not yet run end-to-end

Agent endpoints

Point your agent at:

  • /llms.txt — index of all pages with one-line descriptions (llms.txt v2 spec)
  • /llms-full.txt — every page concatenated into one download
  • /raw/<slug>.md — byte-identical source markdown for any page
  • /index.json — machine-readable index with metadata for every page

Categories

Category Covers
inference-engines llama.cpp, Ollama, vLLM, SGLang, ExLlama, TensorRT-LLM, MLM-LLM, KoboldCpp, Jan, llamafile, GPT4All, LocalAI
models Model families, sizes, licenses, VRAM fit, SWE-bench scores
hardware VRAM math, quantisation formats, KV cache, CPU/AMD/Apple Silicon, buying guidance
application-stack OpenAI-compat serving, agent frameworks, RAG, fine-tuning, eval, speech, image gen
practice-ops Best practices, community, licensing, EU AI Act, security, drivers, cost

Every page carries see_also in front-matter pointing to related pages by slug. The theme renders these as a "Related pages" block at the bottom, plus a generated related-by-tag list. Agents should follow those, not guess URL patterns.

var target=document.getElementById(location.hash.slice(1));target&&target.name&&(target.checked=target.name.startsWith("__tabbed_"))
{"annotate": null, "base": "../../..", "features": [], "search": "../../../assets/javascripts/workers/search.2c215733.min.js", "tags": null, "translations": {"clipboard.copied": "Copied to clipboard", "clipboard.copy": "Copy to clipboard", "search.result.more.one": "1 more on this page", "search.result.more.other": "# more on this page", "search.result.none": "No matching documents", "search.result.one": "1 matching document", "search.result.other": "# matching documents", "search.result.placeholder": "Type to start searching", "search.result.term.missing": "Missing", "select.version": "Select version"}, "version": null}