4028 lines
74 KiB
HTML
4028 lines
74 KiB
HTML
|
||
<!doctype html>
|
||
<html lang="en" class="no-js">
|
||
<head>
|
||
|
||
<meta charset="utf-8">
|
||
<meta name="viewport" content="width=device-width,initial-scale=1">
|
||
|
||
<meta name="description" content="Single source of truth for the homelab">
|
||
|
||
|
||
|
||
<link rel="canonical" href="http://192.168.1.8:13080/services/llm-benchmark/">
|
||
|
||
|
||
<link rel="prev" href="../doc-ingestion/">
|
||
|
||
|
||
<link rel="next" href="../../security/data-leak-audit-comparison/">
|
||
|
||
|
||
|
||
|
||
|
||
<link rel="icon" href="../../assets/images/favicon.png">
|
||
<meta name="generator" content="mkdocs-1.6.1, mkdocs-material-9.7.6">
|
||
|
||
|
||
|
||
<title>LLM benchmark - nuclide.systems docs</title>
|
||
|
||
|
||
|
||
<link rel="stylesheet" href="../../assets/stylesheets/main.484c7ddc.min.css">
|
||
|
||
|
||
<link rel="stylesheet" href="../../assets/stylesheets/palette.ab4e12ef.min.css">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
|
||
<link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Roboto:300,300i,400,400i,700,700i%7CRoboto+Mono:400,400i,700,700i&display=fallback">
|
||
<style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
|
||
|
||
|
||
|
||
<script>__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){}}</script>
|
||
|
||
|
||
|
||
|
||
|
||
</head>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<body dir="ltr" data-md-color-scheme="slate" data-md-color-primary="black" data-md-color-accent="blue">
|
||
|
||
|
||
<input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
|
||
<input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
|
||
<label class="md-overlay" for="__drawer"></label>
|
||
<div data-md-component="skip">
|
||
|
||
|
||
<a href="#llm-model-benchmark-service-catalogue" class="md-skip">
|
||
Skip to content
|
||
</a>
|
||
|
||
</div>
|
||
<div data-md-component="announce">
|
||
|
||
</div>
|
||
|
||
|
||
|
||
|
||
<header class="md-header" data-md-component="header">
|
||
<nav class="md-header__inner md-grid" aria-label="Header">
|
||
<a href="../.." title="nuclide.systems docs" class="md-header__button md-logo" aria-label="nuclide.systems docs" data-md-component="logo">
|
||
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54"/></svg>
|
||
|
||
</a>
|
||
<label class="md-header__button md-icon" for="__drawer">
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3zm0 5h18v2H3zm0 5h18v2H3z"/></svg>
|
||
</label>
|
||
<div class="md-header__title" data-md-component="header-title">
|
||
<div class="md-header__ellipsis">
|
||
<div class="md-header__topic">
|
||
<span class="md-ellipsis">
|
||
nuclide.systems docs
|
||
</span>
|
||
</div>
|
||
<div class="md-header__topic" data-md-component="header-topic">
|
||
<span class="md-ellipsis">
|
||
|
||
LLM benchmark
|
||
|
||
</span>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
<form class="md-header__option" data-md-component="palette">
|
||
|
||
|
||
|
||
|
||
<input class="md-option" data-md-color-media="(prefers-color-scheme: dark)" data-md-color-scheme="slate" data-md-color-primary="black" data-md-color-accent="blue" aria-hidden="true" type="radio" name="__palette" id="__palette_0">
|
||
|
||
|
||
|
||
|
||
|
||
<input class="md-option" data-md-color-media="(prefers-color-scheme: light)" data-md-color-scheme="default" data-md-color-primary="black" data-md-color-accent="blue" aria-hidden="true" type="radio" name="__palette" id="__palette_1">
|
||
|
||
|
||
</form>
|
||
|
||
|
||
|
||
<script>var palette=__md_get("__palette");if(palette&&palette.color){if("(prefers-color-scheme)"===palette.color.media){var media=matchMedia("(prefers-color-scheme: light)"),input=document.querySelector(media.matches?"[data-md-color-media='(prefers-color-scheme: light)']":"[data-md-color-media='(prefers-color-scheme: dark)']");palette.color.media=input.getAttribute("data-md-color-media"),palette.color.scheme=input.getAttribute("data-md-color-scheme"),palette.color.primary=input.getAttribute("data-md-color-primary"),palette.color.accent=input.getAttribute("data-md-color-accent")}for(var[key,value]of Object.entries(palette.color))document.body.setAttribute("data-md-color-"+key,value)}</script>
|
||
|
||
|
||
|
||
|
||
|
||
<label class="md-header__button md-icon" for="__search">
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.52 6.52 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5"/></svg>
|
||
</label>
|
||
<div class="md-search" data-md-component="search" role="dialog">
|
||
<label class="md-search__overlay" for="__search"></label>
|
||
<div class="md-search__inner" role="search">
|
||
<form class="md-search__form" name="search">
|
||
<input type="text" class="md-search__input" name="query" aria-label="Search" placeholder="Search" autocapitalize="off" autocorrect="off" autocomplete="off" spellcheck="false" data-md-component="search-query" required>
|
||
<label class="md-search__icon md-icon" for="__search">
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.52 6.52 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5"/></svg>
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M20 11v2H8l5.5 5.5-1.42 1.42L4.16 12l7.92-7.92L13.5 5.5 8 11z"/></svg>
|
||
</label>
|
||
<nav class="md-search__options" aria-label="Search">
|
||
|
||
<button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M19 6.41 17.59 5 12 10.59 6.41 5 5 6.41 10.59 12 5 17.59 6.41 19 12 13.41 17.59 19 19 17.59 13.41 12z"/></svg>
|
||
</button>
|
||
</nav>
|
||
|
||
<div class="md-search__suggest" data-md-component="search-suggest"></div>
|
||
|
||
</form>
|
||
<div class="md-search__output">
|
||
<div class="md-search__scrollwrap" tabindex="0" data-md-scrollfix>
|
||
<div class="md-search-result" data-md-component="search-result">
|
||
<div class="md-search-result__meta">
|
||
Initializing search
|
||
</div>
|
||
<ol class="md-search-result__list" role="presentation"></ol>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
</nav>
|
||
|
||
</header>
|
||
|
||
<div class="md-container" data-md-component="container">
|
||
|
||
|
||
|
||
|
||
|
||
<nav class="md-tabs" aria-label="Tabs" data-md-component="tabs">
|
||
<div class="md-grid">
|
||
<ul class="md-tabs__list">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../.." class="md-tabs__link">
|
||
|
||
|
||
|
||
|
||
|
||
Home
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../CHANGELOG/" class="md-tabs__link">
|
||
|
||
|
||
|
||
|
||
|
||
Changelog
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../RESUME/" class="md-tabs__link">
|
||
|
||
|
||
|
||
|
||
|
||
Resume
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../ct-inventory/" class="md-tabs__link">
|
||
|
||
|
||
|
||
|
||
|
||
CT inventory
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../infra/proxmox-state/" class="md-tabs__link">
|
||
|
||
|
||
|
||
Infra
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item md-tabs__item--active">
|
||
<a href="../homelab-architecture/" class="md-tabs__link">
|
||
|
||
|
||
|
||
Services
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../security/data-leak-audit-comparison/" class="md-tabs__link">
|
||
|
||
|
||
|
||
Security & audits
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../ideas/stack-ideas/" class="md-tabs__link">
|
||
|
||
|
||
|
||
Ideas
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-tabs__item">
|
||
<a href="../../history/traefik-migration/" class="md-tabs__link">
|
||
|
||
|
||
|
||
History
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
</ul>
|
||
</div>
|
||
</nav>
|
||
|
||
|
||
|
||
<main class="md-main" data-md-component="main">
|
||
<div class="md-main__inner md-grid">
|
||
|
||
|
||
|
||
<div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
|
||
<div class="md-sidebar__scrollwrap">
|
||
<div class="md-sidebar__inner">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<nav class="md-nav md-nav--primary md-nav--lifted" aria-label="Navigation" data-md-level="0">
|
||
<label class="md-nav__title" for="__drawer">
|
||
<a href="../.." title="nuclide.systems docs" class="md-nav__button md-logo" aria-label="nuclide.systems docs" data-md-component="logo">
|
||
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54"/></svg>
|
||
|
||
</a>
|
||
nuclide.systems docs
|
||
</label>
|
||
|
||
<ul class="md-nav__list" data-md-scrollfix>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../.." class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Home
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../CHANGELOG/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Changelog
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../RESUME/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Resume
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../ct-inventory/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
CT inventory
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item md-nav__item--nested">
|
||
|
||
|
||
|
||
|
||
|
||
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_5" >
|
||
|
||
|
||
<label class="md-nav__link" for="__nav_5" id="__nav_5_label" tabindex="0">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Infra
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
<span class="md-nav__icon md-icon"></span>
|
||
</label>
|
||
|
||
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_5_label" aria-expanded="false">
|
||
<label class="md-nav__title" for="__nav_5">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
|
||
|
||
Infra
|
||
|
||
|
||
</label>
|
||
<ul class="md-nav__list" data-md-scrollfix>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/proxmox-state/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Proxmox state
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/portmap/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Port map
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/storage/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Storage
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/volumes/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Volumes
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/docker-networks/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Docker networks
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/connection-hosts/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Connection hosts
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../infra/proxmox-memory-audit/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Memory audit
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item md-nav__item--active md-nav__item--section md-nav__item--nested">
|
||
|
||
|
||
|
||
<input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_6" checked>
|
||
|
||
|
||
<label class="md-nav__link" for="__nav_6" id="__nav_6_label" tabindex="">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Services
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
<span class="md-nav__icon md-icon"></span>
|
||
</label>
|
||
|
||
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_6_label" aria-expanded="true">
|
||
<label class="md-nav__title" for="__nav_6">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
|
||
|
||
Services
|
||
|
||
|
||
</label>
|
||
<ul class="md-nav__list" data-md-scrollfix>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../homelab-architecture/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Homelab architecture
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../dev-environment/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Dev environment (Coder + Gitea)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../mcp-gateway/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
MCP gateway
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../mcp-servers/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
MCP servers
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../pocket-id/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Pocket-ID (OIDC)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../backrest/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Backrest (backups)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../databases/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Databases
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../secrets-manager/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Secrets manager
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../comfyui/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
ComfyUI
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../zoraxy/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Zoraxy
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../adguard-dns/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
AdGuard DNS
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../cloud-gpu/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Cloud GPU
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../arcane/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Arcane
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../doc-ingestion/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Doc ingestion pipeline
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item md-nav__item--active">
|
||
|
||
<input class="md-nav__toggle md-toggle" type="checkbox" id="__toc">
|
||
|
||
|
||
|
||
|
||
|
||
<label class="md-nav__link md-nav__link--active" for="__toc">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
LLM benchmark
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
<span class="md-nav__icon md-icon"></span>
|
||
</label>
|
||
|
||
<a href="./" class="md-nav__link md-nav__link--active">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
LLM benchmark
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
|
||
|
||
|
||
<nav class="md-nav md-nav--secondary" aria-label="Table of contents">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<label class="md-nav__title" for="__toc">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
Table of contents
|
||
</label>
|
||
<ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#overview" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Overview
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#performance-tiers" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Performance Tiers
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#model-catalogue" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Model Catalogue
|
||
|
||
</span>
|
||
</a>
|
||
|
||
<nav class="md-nav" aria-label="Model Catalogue">
|
||
<ul class="md-nav__list">
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#chat-reasoning-models" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Chat & Reasoning Models
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#embedding-models" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Embedding Models
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#audio-models-tts-asr" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Audio Models (TTS / ASR)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#image-models" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Image Models
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#service-recommendations" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Service Recommendations
|
||
|
||
</span>
|
||
</a>
|
||
|
||
<nav class="md-nav" aria-label="Service Recommendations">
|
||
<ul class="md-nav__list">
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#nextcloud-assistant" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Nextcloud Assistant
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#karakeep-bookmarks-reading" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Karakeep (Bookmarks / Reading)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#home-assistant" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Home Assistant
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#lobechat-default" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
LobeChat Default
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#code-assistant-coder-ide" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Code Assistant (Coder / IDE)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#document-ocr-ingestion" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Document OCR / Ingestion
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#embeddings-karakeep-lobechat-kb" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Embeddings (Karakeep / LobeChat KB)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#image-generation" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Image Generation
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#tts-voice-interfaces" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
TTS (Voice Interfaces)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#transcription-meetings-voice" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Transcription (Meetings / Voice)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#reasoning-analysis" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Reasoning / Analysis
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#batch-offline-processing" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Batch / Offline Processing
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#aliases-duplicates" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Aliases & Duplicates
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#experimental-not-yet-validated" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Experimental / Not Yet Validated
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#raw-benchmark-data" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Raw Benchmark Data
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
|
||
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item md-nav__item--nested">
|
||
|
||
|
||
|
||
|
||
|
||
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_7" >
|
||
|
||
|
||
<label class="md-nav__link" for="__nav_7" id="__nav_7_label" tabindex="0">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Security & audits
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
<span class="md-nav__icon md-icon"></span>
|
||
</label>
|
||
|
||
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_7_label" aria-expanded="false">
|
||
<label class="md-nav__title" for="__nav_7">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
|
||
|
||
Security & audits
|
||
|
||
|
||
</label>
|
||
<ul class="md-nav__list" data-md-scrollfix>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../security/data-leak-audit-comparison/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Comparison
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../security/data-leak-audit-2026-05-20-tr004-cloud-sandbox/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
2026-05-20 · cloud-sandbox breach
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../security/data-leak-audit-2026-05-21-tr004-artifacts/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
2026-05-21 · artifacts (clean)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../security/audit-claude-code-meta/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Self-audit (Claude Code)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../security/transcripts/audit-chat-2026-05-21/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Transcript (audit session)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item md-nav__item--nested">
|
||
|
||
|
||
|
||
|
||
|
||
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_8" >
|
||
|
||
|
||
<label class="md-nav__link" for="__nav_8" id="__nav_8_label" tabindex="0">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Ideas
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
<span class="md-nav__icon md-icon"></span>
|
||
</label>
|
||
|
||
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_8_label" aria-expanded="false">
|
||
<label class="md-nav__title" for="__nav_8">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
|
||
|
||
Ideas
|
||
|
||
|
||
</label>
|
||
<ul class="md-nav__list" data-md-scrollfix>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../ideas/stack-ideas/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Stack ideas
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item md-nav__item--nested">
|
||
|
||
|
||
|
||
|
||
|
||
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_9" >
|
||
|
||
|
||
<label class="md-nav__link" for="__nav_9" id="__nav_9_label" tabindex="0">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
History
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
<span class="md-nav__icon md-icon"></span>
|
||
</label>
|
||
|
||
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_9_label" aria-expanded="false">
|
||
<label class="md-nav__title" for="__nav_9">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
|
||
|
||
History
|
||
|
||
|
||
</label>
|
||
<ul class="md-nav__list" data-md-scrollfix>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../history/traefik-migration/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Traefik (abandoned 2026-05-16)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../history/traefik-migration-docker-labels/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Traefik labels (abandoned)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../history/mcp-gateway-requirements/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
MCP gateway requirements (superseded)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../history/scrubbing-list-2026-05-17/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Scrubbing list (2026-05-17)
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<li class="md-nav__item">
|
||
<a href="../../history/case-study/" class="md-nav__link">
|
||
|
||
|
||
|
||
<span class="md-ellipsis">
|
||
|
||
|
||
Case study
|
||
|
||
|
||
|
||
</span>
|
||
|
||
|
||
|
||
</a>
|
||
</li>
|
||
|
||
|
||
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
|
||
|
||
</ul>
|
||
</nav>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
<div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
|
||
<div class="md-sidebar__scrollwrap">
|
||
<div class="md-sidebar__inner">
|
||
|
||
|
||
<nav class="md-nav md-nav--secondary" aria-label="Table of contents">
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<label class="md-nav__title" for="__toc">
|
||
<span class="md-nav__icon md-icon"></span>
|
||
Table of contents
|
||
</label>
|
||
<ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#overview" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Overview
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#performance-tiers" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Performance Tiers
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#model-catalogue" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Model Catalogue
|
||
|
||
</span>
|
||
</a>
|
||
|
||
<nav class="md-nav" aria-label="Model Catalogue">
|
||
<ul class="md-nav__list">
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#chat-reasoning-models" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Chat & Reasoning Models
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#embedding-models" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Embedding Models
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#audio-models-tts-asr" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Audio Models (TTS / ASR)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#image-models" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Image Models
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#service-recommendations" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Service Recommendations
|
||
|
||
</span>
|
||
</a>
|
||
|
||
<nav class="md-nav" aria-label="Service Recommendations">
|
||
<ul class="md-nav__list">
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#nextcloud-assistant" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Nextcloud Assistant
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#karakeep-bookmarks-reading" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Karakeep (Bookmarks / Reading)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#home-assistant" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Home Assistant
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#lobechat-default" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
LobeChat Default
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#code-assistant-coder-ide" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Code Assistant (Coder / IDE)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#document-ocr-ingestion" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Document OCR / Ingestion
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#embeddings-karakeep-lobechat-kb" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Embeddings (Karakeep / LobeChat KB)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#image-generation" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Image Generation
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#tts-voice-interfaces" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
TTS (Voice Interfaces)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#transcription-meetings-voice" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Transcription (Meetings / Voice)
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#reasoning-analysis" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Reasoning / Analysis
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#batch-offline-processing" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Batch / Offline Processing
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
</nav>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#aliases-duplicates" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Aliases & Duplicates
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#experimental-not-yet-validated" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Experimental / Not Yet Validated
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
<li class="md-nav__item">
|
||
<a href="#raw-benchmark-data" class="md-nav__link">
|
||
<span class="md-ellipsis">
|
||
|
||
Raw Benchmark Data
|
||
|
||
</span>
|
||
</a>
|
||
|
||
</li>
|
||
|
||
</ul>
|
||
|
||
</nav>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
|
||
<div class="md-content" data-md-component="content">
|
||
|
||
<article class="md-content__inner md-typeset">
|
||
|
||
|
||
|
||
|
||
|
||
<h1 id="llm-model-benchmark-service-catalogue">LLM Model Benchmark & Service Catalogue<a class="headerlink" href="#llm-model-benchmark-service-catalogue" title="Permanent link">¶</a></h1>
|
||
<p><em>Last updated: 2026-05-23 | Models: 58 | Source: LiteLLM /model/info + live benchmarks</em></p>
|
||
<h2 id="overview">Overview<a class="headerlink" href="#overview" title="Permanent link">¶</a></h2>
|
||
<p>This catalogue covers all 60 models registered in the homelab LiteLLM proxy (<code>http://192.168.1.40:14000</code>).
|
||
Live latency figures are TTFT proxies measured from this host via a single <code>max_tokens=5</code> completion request.
|
||
Speed tiers are based on live measurements and published inference benchmarks.</p>
|
||
<p><strong>Providers at a glance:</strong></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Provider</th>
|
||
<th>Models</th>
|
||
<th>Notes</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td>Claude (Anthropic via openai-compat)</td>
|
||
<td>6</td>
|
||
<td>claude-max subscription; temperature=0.7; aliases included</td>
|
||
</tr>
|
||
<tr>
|
||
<td>Mistral API</td>
|
||
<td>10</td>
|
||
<td>voxtral voice family + codestral + OCR</td>
|
||
</tr>
|
||
<tr>
|
||
<td>Gemini API</td>
|
||
<td>12</td>
|
||
<td>Flash/Pro/embedding families</td>
|
||
</tr>
|
||
<tr>
|
||
<td>SAIA (self-hosted GPU cluster)</td>
|
||
<td>22</td>
|
||
<td>OpenAI-compatible; local GPU inference</td>
|
||
</tr>
|
||
<tr>
|
||
<td>Groq</td>
|
||
<td>2</td>
|
||
<td>Ultra-fast cloud inference</td>
|
||
</tr>
|
||
<tr>
|
||
<td>Cerebras</td>
|
||
<td>2</td>
|
||
<td>Ultra-fast wafer-scale inference</td>
|
||
</tr>
|
||
<tr>
|
||
<td>Cohere</td>
|
||
<td>2</td>
|
||
<td>Embeddings only</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<hr />
|
||
<h2 id="performance-tiers">Performance Tiers<a class="headerlink" href="#performance-tiers" title="Permanent link">¶</a></h2>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Tier</th>
|
||
<th>Symbol</th>
|
||
<th>Typical TTFT</th>
|
||
<th>Profile</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Ultra-fast</strong></td>
|
||
<td>🚀</td>
|
||
<td>< 200 ms</td>
|
||
<td>Groq, Cerebras, cached SAIA small models</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fast</strong></td>
|
||
<td>⚡</td>
|
||
<td>200–600 ms</td>
|
||
<td>Mistral API, Gemini Flash, SAIA mid-size</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Standard</strong></td>
|
||
<td>🔵</td>
|
||
<td>600–2 000 ms</td>
|
||
<td>Claude, Gemini Pro, large API models</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Self-hosted</strong></td>
|
||
<td>🏠</td>
|
||
<td>varies</td>
|
||
<td>SAIA cluster; latency depends on GPU load & model size</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<blockquote>
|
||
<p>Note: SAIA models with very low latency (< 50 ms) on the trivial benchmark likely hit a cached/KV-prefilled response; real-world TTFT for longer prompts will be higher. Treat SAIA figures as best-case.</p>
|
||
</blockquote>
|
||
<hr />
|
||
<h2 id="model-catalogue">Model Catalogue<a class="headerlink" href="#model-catalogue" title="Permanent link">¶</a></h2>
|
||
<h3 id="chat-reasoning-models">Chat & Reasoning Models<a class="headerlink" href="#chat-reasoning-models" title="Permanent link">¶</a></h3>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Model ID</th>
|
||
<th>Provider</th>
|
||
<th>Backend</th>
|
||
<th>Context</th>
|
||
<th>Vision</th>
|
||
<th>Tools</th>
|
||
<th>Cost In $/1M</th>
|
||
<th>Cost Out $/1M</th>
|
||
<th>Live Latency</th>
|
||
<th>Speed Tier</th>
|
||
<th>Notes</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td>claude-sonnet-4-6</td>
|
||
<td>Anthropic</td>
|
||
<td>openai-compat</td>
|
||
<td>200K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$3.00</td>
|
||
<td>$15.00</td>
|
||
<td>2 083 ms</td>
|
||
<td>🔵</td>
|
||
<td>Flagship; temp=0.7</td>
|
||
</tr>
|
||
<tr>
|
||
<td>claude-opus-4-7</td>
|
||
<td>Anthropic</td>
|
||
<td>openai-compat</td>
|
||
<td>200K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$15.00</td>
|
||
<td>$75.00</td>
|
||
<td>3 218 ms</td>
|
||
<td>🔵</td>
|
||
<td>Highest capability; temp=0.7</td>
|
||
</tr>
|
||
<tr>
|
||
<td>claude-haiku-4-5</td>
|
||
<td>Anthropic</td>
|
||
<td>openai-compat</td>
|
||
<td>200K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$0.80</td>
|
||
<td>$4.00</td>
|
||
<td>1 453 ms</td>
|
||
<td>🔵</td>
|
||
<td>Fast + cheap; temp=0.7</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-small-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>256K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>160 ms</td>
|
||
<td>🚀</td>
|
||
<td>Voice+text multimodal</td>
|
||
</tr>
|
||
<tr>
|
||
<td>mistral-small-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>131K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$0.06</td>
|
||
<td>$0.18</td>
|
||
<td>222 ms</td>
|
||
<td>🚀</td>
|
||
<td>Cheapest Mistral chat</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-mini-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>100K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>ERROR</td>
|
||
<td>❌</td>
|
||
<td>Proxy config error (saia-image-proxy unreachable)</td>
|
||
</tr>
|
||
<tr>
|
||
<td>codestral-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>16K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$1.00</td>
|
||
<td>$3.00</td>
|
||
<td>275 ms</td>
|
||
<td>⚡</td>
|
||
<td>Coding-specialised Mistral</td>
|
||
</tr>
|
||
<tr>
|
||
<td>mistral-large-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>262K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$0.50</td>
|
||
<td>$1.50</td>
|
||
<td>309 ms</td>
|
||
<td>⚡</td>
|
||
<td>Flagship Mistral chat</td>
|
||
</tr>
|
||
<tr>
|
||
<td>pixtral-large-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>128K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$2.00</td>
|
||
<td>$6.00</td>
|
||
<td>27 ms</td>
|
||
<td>🚀</td>
|
||
<td>Vision flagship; very low latency (likely cached)</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-mini-realtime-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>4K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>ERROR</td>
|
||
<td>❌</td>
|
||
<td>Invalid model per API; realtime/ws endpoint only</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-flash</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$0.30</td>
|
||
<td>$2.50</td>
|
||
<td>501 ms</td>
|
||
<td>⚡</td>
|
||
<td>Best-value Gemini; fast + smart</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-flash-lite</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$0.10</td>
|
||
<td>$0.40</td>
|
||
<td>518 ms</td>
|
||
<td>⚡</td>
|
||
<td>Lightest + cheapest Gemini</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-pro</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$1.25</td>
|
||
<td>$10.00</td>
|
||
<td>888 ms</td>
|
||
<td>🔵</td>
|
||
<td>Top Gemini reasoning</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-3.1-pro-preview</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$2.00</td>
|
||
<td>$12.00</td>
|
||
<td>1 215 ms</td>
|
||
<td>🔵</td>
|
||
<td>Next-gen Gemini Pro preview</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-3-pro-preview</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$2.00</td>
|
||
<td>$12.00</td>
|
||
<td>1 258 ms</td>
|
||
<td>🔵</td>
|
||
<td>Gemini 3 Pro preview</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-3.1-flash-lite</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>—</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>482 ms</td>
|
||
<td>⚡</td>
|
||
<td>Gemini 3.1 flash lite preview</td>
|
||
</tr>
|
||
<tr>
|
||
<td>devstral-2-123b-instruct-2512</td>
|
||
<td>Mistral</td>
|
||
<td>SAIA</td>
|
||
<td>4K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>157 ms</td>
|
||
<td>🏠</td>
|
||
<td>123B coding model via SAIA GPU</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3-coder-30b-a3b-instruct</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>32K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>5 ms</td>
|
||
<td>🏠</td>
|
||
<td>MoE coding model; 5 ms = cached</td>
|
||
</tr>
|
||
<tr>
|
||
<td>openai-gpt-oss-120b</td>
|
||
<td>OpenAI</td>
|
||
<td>SAIA</td>
|
||
<td>131K</td>
|
||
<td>✗</td>
|
||
<td>✗</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>118 ms</td>
|
||
<td>🏠</td>
|
||
<td>OpenAI open-weight 120B via SAIA</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3-omni-30b-a3b-instruct</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>16K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>178 ms</td>
|
||
<td>🏠</td>
|
||
<td>Multimodal MoE; voice+vision</td>
|
||
</tr>
|
||
<tr>
|
||
<td>llama-3.3-70b-instruct</td>
|
||
<td>Meta</td>
|
||
<td>SAIA</td>
|
||
<td>131K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>168 ms</td>
|
||
<td>🏠</td>
|
||
<td>Reliable general-purpose 70B</td>
|
||
</tr>
|
||
<tr>
|
||
<td>deepseek-r1-distill-llama-70b</td>
|
||
<td>DeepSeek</td>
|
||
<td>SAIA</td>
|
||
<td>131K</td>
|
||
<td>✗</td>
|
||
<td>✗</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>325 ms</td>
|
||
<td>🏠</td>
|
||
<td>R1 reasoning distill; outputs <code><think></code> tokens</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-35b-a3b</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>250 ms</td>
|
||
<td>🏠</td>
|
||
<td>MoE 35B A3B</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-27b</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>131K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>266 ms</td>
|
||
<td>🏠</td>
|
||
<td>Dense 27B</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.6-35b-a3b</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>204 ms</td>
|
||
<td>🏠</td>
|
||
<td>MoE 35B A3B v3.6</td>
|
||
</tr>
|
||
<tr>
|
||
<td>apertus-70b-instruct-2509</td>
|
||
<td>Apertus</td>
|
||
<td>SAIA</td>
|
||
<td>4K</td>
|
||
<td>✗</td>
|
||
<td>✗</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>143 ms</td>
|
||
<td>🏠</td>
|
||
<td>Small context; general chat</td>
|
||
</tr>
|
||
<tr>
|
||
<td>glm-4.7</td>
|
||
<td>Zhipu</td>
|
||
<td>SAIA</td>
|
||
<td>128K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>210 ms</td>
|
||
<td>🏠</td>
|
||
<td>GLM-4 series</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3-30b-a3b-instruct-2507</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>262K</td>
|
||
<td>✗</td>
|
||
<td>✗</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>127 ms</td>
|
||
<td>🏠</td>
|
||
<td>MoE 30B, very large context</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemma-3-27b-it</td>
|
||
<td>Google</td>
|
||
<td>SAIA</td>
|
||
<td>131K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>284 ms</td>
|
||
<td>🏠</td>
|
||
<td>Gemma 3 27B instruct</td>
|
||
</tr>
|
||
<tr>
|
||
<td>internvl3.5-30b-a3b</td>
|
||
<td>InternLM</td>
|
||
<td>SAIA</td>
|
||
<td>16K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>116 ms</td>
|
||
<td>🏠</td>
|
||
<td>Vision+tools MoE 30B</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-122b-a10b</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>372 ms</td>
|
||
<td>🏠</td>
|
||
<td>MoE 122B A10B; larger/slower</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-397b-a17b</td>
|
||
<td>Alibaba</td>
|
||
<td>SAIA</td>
|
||
<td>65K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>248 ms</td>
|
||
<td>🏠</td>
|
||
<td>Largest SAIA MoE model</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemma-4-31b-it</td>
|
||
<td>Google</td>
|
||
<td>SAIA</td>
|
||
<td>8K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>177 ms</td>
|
||
<td>🏠</td>
|
||
<td>Gemma 4 multimodal</td>
|
||
</tr>
|
||
<tr>
|
||
<td>meta-llama/llama-4-scout-17b-16e-instruct</td>
|
||
<td>Meta</td>
|
||
<td>Groq</td>
|
||
<td>8K</td>
|
||
<td>✓</td>
|
||
<td>✓</td>
|
||
<td>$0.11</td>
|
||
<td>$0.34</td>
|
||
<td>442 ms</td>
|
||
<td>🚀</td>
|
||
<td>Groq-accelerated; 442 ms incl. queue</td>
|
||
</tr>
|
||
<tr>
|
||
<td>cerebras-llama-3.1-8b</td>
|
||
<td>Meta</td>
|
||
<td>Cerebras</td>
|
||
<td>128K</td>
|
||
<td>✗</td>
|
||
<td>✓</td>
|
||
<td>$0.10</td>
|
||
<td>$0.10</td>
|
||
<td>258 ms</td>
|
||
<td>🚀</td>
|
||
<td>Wafer-scale ~2 000 TPS</td>
|
||
</tr>
|
||
<tr>
|
||
<td>cerebras-qwen-3-235b</td>
|
||
<td>Alibaba</td>
|
||
<td>Cerebras</td>
|
||
<td>—</td>
|
||
<td>✗</td>
|
||
<td>✗</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
<td>199 ms</td>
|
||
<td>🚀</td>
|
||
<td>235B at wafer-scale speed</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<h3 id="embedding-models">Embedding Models<a class="headerlink" href="#embedding-models" title="Permanent link">¶</a></h3>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Model ID</th>
|
||
<th>Provider</th>
|
||
<th>Backend</th>
|
||
<th>Context</th>
|
||
<th>Dimensions</th>
|
||
<th>Cost $/1M</th>
|
||
<th>Notes</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td>gemini-embedding-2</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>8K</td>
|
||
<td>—</td>
|
||
<td>$0.20</td>
|
||
<td>Primary Gemini embedding</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-embedding-001</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>2K</td>
|
||
<td>—</td>
|
||
<td>$0.15</td>
|
||
<td>Legacy Gemini embedding</td>
|
||
</tr>
|
||
<tr>
|
||
<td>text-embedding-ada-002</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>8K</td>
|
||
<td>—</td>
|
||
<td>$0.20</td>
|
||
<td>Alias → gemini-embedding-2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>text-embedding-3-small</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>8K</td>
|
||
<td>—</td>
|
||
<td>$0.20</td>
|
||
<td>Alias → gemini-embedding-2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>text-embedding-3-large</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>8K</td>
|
||
<td>—</td>
|
||
<td>$0.20</td>
|
||
<td>Alias → gemini-embedding-2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>multilingual-e5-large-instruct</td>
|
||
<td>Microsoft</td>
|
||
<td>SAIA</td>
|
||
<td>8K</td>
|
||
<td>1 024</td>
|
||
<td>—</td>
|
||
<td>Self-hosted multilingual; strong for DE/EN RAG</td>
|
||
</tr>
|
||
<tr>
|
||
<td>cohere-embed-multilingual-v3</td>
|
||
<td>Cohere</td>
|
||
<td>Cohere API</td>
|
||
<td>1K</td>
|
||
<td>1 024</td>
|
||
<td>$0.10</td>
|
||
<td>100+ languages</td>
|
||
</tr>
|
||
<tr>
|
||
<td>cohere-embed-english-v3</td>
|
||
<td>Cohere</td>
|
||
<td>Cohere API</td>
|
||
<td>1K</td>
|
||
<td>1 024</td>
|
||
<td>$0.10</td>
|
||
<td>English-only; higher EN accuracy</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<h3 id="audio-models-tts-asr">Audio Models (TTS / ASR)<a class="headerlink" href="#audio-models-tts-asr" title="Permanent link">¶</a></h3>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Model ID</th>
|
||
<th>Provider</th>
|
||
<th>Backend</th>
|
||
<th>Type</th>
|
||
<th>Language</th>
|
||
<th>Notes</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td>tts-1-de</td>
|
||
<td>SAIA</td>
|
||
<td>Piper</td>
|
||
<td>TTS</td>
|
||
<td>German</td>
|
||
<td>Self-hosted German TTS</td>
|
||
</tr>
|
||
<tr>
|
||
<td>tts-1</td>
|
||
<td>SAIA</td>
|
||
<td>Kokoro-82M</td>
|
||
<td>TTS</td>
|
||
<td>EN + others</td>
|
||
<td>Self-hosted multilingual TTS</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-mini-tts-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>TTS</td>
|
||
<td>Multilingual</td>
|
||
<td>Mistral voice synthesis</td>
|
||
</tr>
|
||
<tr>
|
||
<td>saia-whisper</td>
|
||
<td>SAIA</td>
|
||
<td>whisper-large-v2</td>
|
||
<td>ASR</td>
|
||
<td>Multilingual</td>
|
||
<td>Self-hosted transcription</td>
|
||
</tr>
|
||
<tr>
|
||
<td>whisper-1</td>
|
||
<td>SAIA</td>
|
||
<td>faster-whisper-large-v3</td>
|
||
<td>ASR</td>
|
||
<td>Multilingual</td>
|
||
<td>Faster self-hosted transcription</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-mini-transcribe-2507</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>ASR</td>
|
||
<td>Multilingual</td>
|
||
<td>Mistral audio transcription; ctx 16K</td>
|
||
</tr>
|
||
<tr>
|
||
<td>whisper-large-v3-turbo</td>
|
||
<td>Meta</td>
|
||
<td>Groq</td>
|
||
<td>ASR</td>
|
||
<td>Multilingual</td>
|
||
<td>Groq-accelerated; fastest transcription</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<h3 id="image-models">Image Models<a class="headerlink" href="#image-models" title="Permanent link">¶</a></h3>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Model ID</th>
|
||
<th>Provider</th>
|
||
<th>Backend</th>
|
||
<th>Type</th>
|
||
<th>Notes</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td>saia-flux</td>
|
||
<td>SAIA</td>
|
||
<td>FLUX</td>
|
||
<td>Image gen</td>
|
||
<td>Self-hosted FLUX; note: garbles text labels</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-flash-image</td>
|
||
<td>Google</td>
|
||
<td>Gemini API</td>
|
||
<td>Image gen</td>
|
||
<td>ctx 32K; multimodal image generation</td>
|
||
</tr>
|
||
<tr>
|
||
<td>saia-image-edit</td>
|
||
<td>SAIA</td>
|
||
<td>Qwen-Image-Edit</td>
|
||
<td>Image edit</td>
|
||
<td>Image editing/inpainting</td>
|
||
</tr>
|
||
<tr>
|
||
<td>mistral-ocr-latest</td>
|
||
<td>Mistral</td>
|
||
<td>Mistral API</td>
|
||
<td>OCR</td>
|
||
<td>Document OCR; not a chat model</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<hr />
|
||
<h2 id="service-recommendations">Service Recommendations<a class="headerlink" href="#service-recommendations" title="Permanent link">¶</a></h2>
|
||
<h3 id="nextcloud-assistant">Nextcloud Assistant<a class="headerlink" href="#nextcloud-assistant" title="Permanent link">¶</a></h3>
|
||
<p><em>Smart file/email/calendar assistant, summaries, writing help — multilingual DE/EN</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>mistral-small-latest</code></td>
|
||
<td>Cheapest API model with tool use, vision, 131K context, and 222 ms TTFT. Handles German natively.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fallback</strong></td>
|
||
<td><code>claude-haiku-4-5</code></td>
|
||
<td>If higher quality needed; still cost-effective at $0.80/$4 and 1 453 ms TTFT.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Alt (local)</strong></td>
|
||
<td><code>qwen3.5-27b</code></td>
|
||
<td>Free if staying fully on SAIA GPU cluster; 131K context + tools.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-0-1" name="__codelineno-0-1" href="#__codelineno-0-1"></a>model: mistral-small-latest
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="karakeep-bookmarks-reading">Karakeep (Bookmarks / Reading)<a class="headerlink" href="#karakeep-bookmarks-reading" title="Permanent link">¶</a></h3>
|
||
<p><em>Summarise articles, extract key points, tag/categorise — no vision required</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>gemini-2.5-flash-lite</code></td>
|
||
<td>Cheapest API model at $0.10/$0.40, 518 ms TTFT, strong comprehension.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fallback</strong></td>
|
||
<td><code>mistral-small-latest</code></td>
|
||
<td>Slightly pricier but faster at 222 ms.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-1-1" name="__codelineno-1-1" href="#__codelineno-1-1"></a>model: gemini-2.5-flash-lite
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="home-assistant">Home Assistant<a class="headerlink" href="#home-assistant" title="Permanent link">¶</a></h3>
|
||
<p><em>Intent recognition, automation triggers, voice pipeline — ultra-low latency critical</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>cerebras-llama-3.1-8b</code></td>
|
||
<td>258 ms measured TTFT, ~2 000 TPS on Cerebras wafer silicon; best latency for real-time voice. 128K context, tools.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fallback</strong></td>
|
||
<td><code>mistral-small-latest</code></td>
|
||
<td>222 ms TTFT, API-based, reliable tool calling.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Local alt</strong></td>
|
||
<td><code>internvl3.5-30b-a3b</code></td>
|
||
<td>116 ms on SAIA; avoids API cost for high-frequency automations.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-2-1" name="__codelineno-2-1" href="#__codelineno-2-1"></a>model: cerebras-llama-3.1-8b
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="lobechat-default">LobeChat Default<a class="headerlink" href="#lobechat-default" title="Permanent link">¶</a></h3>
|
||
<p><em>General chat assistant for daily use — balanced quality / speed / cost, vision nice</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>gemini-2.5-flash</code></td>
|
||
<td>501 ms, vision, tools, large context, excellent reasoning at $0.30/$2.50. Best all-rounder.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fallback</strong></td>
|
||
<td><code>claude-sonnet-4-6</code></td>
|
||
<td>Higher quality ceiling; use when depth matters over cost.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Free alt</strong></td>
|
||
<td><code>qwen3.5-397b-a17b</code></td>
|
||
<td>Largest self-hosted model; free on SAIA with vision + tools at 248 ms.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-3-1" name="__codelineno-3-1" href="#__codelineno-3-1"></a>model: gemini-2.5-flash
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="code-assistant-coder-ide">Code Assistant (Coder / IDE)<a class="headerlink" href="#code-assistant-coder-ide" title="Permanent link">¶</a></h3>
|
||
<p><em>Code completion, review, debugging — strong code ability, large context, tools</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>claude-sonnet-4-6</code></td>
|
||
<td>Best overall coding + reasoning; 200K context, tool use, reliable output.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fast/cheap</strong></td>
|
||
<td><code>codestral-latest</code></td>
|
||
<td>Coding-specialist Mistral at 275 ms with 16K context; good for completion.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Local coding</strong></td>
|
||
<td><code>devstral-2-123b-instruct-2512</code></td>
|
||
<td>123B SAIA coding model at 157 ms; free inference.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>MoE coding</strong></td>
|
||
<td><code>qwen3-coder-30b-a3b-instruct</code></td>
|
||
<td>32K context, tools, extremely fast (5 ms cached); best SAIA coding model.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-4-1" name="__codelineno-4-1" href="#__codelineno-4-1"></a>model: claude-sonnet-4-6 # IDE / review
|
||
<a id="__codelineno-4-2" name="__codelineno-4-2" href="#__codelineno-4-2"></a>model: qwen3-coder-30b-a3b-instruct # local completion
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="document-ocr-ingestion">Document OCR / Ingestion<a class="headerlink" href="#document-ocr-ingestion" title="Permanent link">¶</a></h3>
|
||
<p><em>Paperless → Docling → extract text — vision + OCR capable, large context</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>mistral-ocr-latest</code></td>
|
||
<td>Dedicated OCR endpoint; purpose-built for document text extraction.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fallback</strong></td>
|
||
<td><code>gemini-2.5-pro</code></td>
|
||
<td>65K context, vision, strong at structured extraction from images.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Alt vision</strong></td>
|
||
<td><code>pixtral-large-latest</code></td>
|
||
<td>Mistral vision flagship at 27 ms (cached); good document parsing.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-5-1" name="__codelineno-5-1" href="#__codelineno-5-1"></a>model: mistral-ocr-latest # OCR pipeline
|
||
<a id="__codelineno-5-2" name="__codelineno-5-2" href="#__codelineno-5-2"></a>model: gemini-2.5-pro # fallback / complex layouts
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="embeddings-karakeep-lobechat-kb">Embeddings (Karakeep / LobeChat KB)<a class="headerlink" href="#embeddings-karakeep-lobechat-kb" title="Permanent link">¶</a></h3>
|
||
<p><em>Semantic search, RAG, knowledge base — multilingual, high dimensions</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary (API)</strong></td>
|
||
<td><code>gemini-embedding-2</code></td>
|
||
<td>8K context, $0.20/1M, strong multilingual.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Primary (local)</strong></td>
|
||
<td><code>multilingual-e5-large-instruct</code></td>
|
||
<td>Self-hosted on SAIA, 1 024-dim, excellent DE/EN RAG, zero API cost.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Multilingual API</strong></td>
|
||
<td><code>cohere-embed-multilingual-v3</code></td>
|
||
<td>100+ languages, 1K context, $0.10/1M.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-6-1" name="__codelineno-6-1" href="#__codelineno-6-1"></a>model: multilingual-e5-large-instruct # local RAG
|
||
<a id="__codelineno-6-2" name="__codelineno-6-2" href="#__codelineno-6-2"></a>model: gemini-embedding-2 # API fallback
|
||
</code></pre></div>
|
||
<blockquote>
|
||
<p><strong>Note:</strong> <code>text-embedding-ada-002</code>, <code>text-embedding-3-small</code>, and <code>text-embedding-3-large</code> are all aliases for <code>gemini-embedding-2</code> — use the canonical ID to avoid confusion.</p>
|
||
</blockquote>
|
||
<hr />
|
||
<h3 id="image-generation">Image Generation<a class="headerlink" href="#image-generation" title="Permanent link">¶</a></h3>
|
||
<p><em>ComfyUI complement, quick drafts</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>saia-flux</code></td>
|
||
<td>Self-hosted FLUX on SAIA GPU; no API cost. Note: avoid text in generated images (garbles).</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>API alt</strong></td>
|
||
<td><code>gemini-2.5-flash-image</code></td>
|
||
<td>Gemini multimodal image gen for quick API-based drafts.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Editing</strong></td>
|
||
<td><code>saia-image-edit</code></td>
|
||
<td>Qwen image editing for inpainting / modifications.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-7-1" name="__codelineno-7-1" href="#__codelineno-7-1"></a>model: saia-flux
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="tts-voice-interfaces">TTS (Voice Interfaces)<a class="headerlink" href="#tts-voice-interfaces" title="Permanent link">¶</a></h3>
|
||
<p><em>Read content aloud, voice responses</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>German</strong></td>
|
||
<td><code>tts-1-de</code></td>
|
||
<td>Self-hosted Piper; native German pronunciation.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Multilingual</strong></td>
|
||
<td><code>tts-1</code></td>
|
||
<td>Self-hosted Kokoro-82M; covers EN + others, zero cost.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>API quality</strong></td>
|
||
<td><code>voxtral-mini-tts-latest</code></td>
|
||
<td>Mistral neural TTS for higher-quality voice synthesis.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-8-1" name="__codelineno-8-1" href="#__codelineno-8-1"></a>model: tts-1-de # German HA / Nextcloud voice
|
||
<a id="__codelineno-8-2" name="__codelineno-8-2" href="#__codelineno-8-2"></a>model: tts-1 # English / multilingual
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="transcription-meetings-voice">Transcription (Meetings / Voice)<a class="headerlink" href="#transcription-meetings-voice" title="Permanent link">¶</a></h3>
|
||
<p><em>Speech to text</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>whisper-large-v3-turbo</code></td>
|
||
<td>Groq-accelerated; fastest available transcription.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Local</strong></td>
|
||
<td><code>whisper-1</code></td>
|
||
<td>faster-whisper-large-v3 on SAIA; fully self-hosted, no API cost.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fallback</strong></td>
|
||
<td><code>saia-whisper</code></td>
|
||
<td>whisper-large-v2 on SAIA; slightly older model.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-9-1" name="__codelineno-9-1" href="#__codelineno-9-1"></a>model: whisper-large-v3-turbo # real-time meetings
|
||
<a id="__codelineno-9-2" name="__codelineno-9-2" href="#__codelineno-9-2"></a>model: whisper-1 # offline / batch
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="reasoning-analysis">Reasoning / Analysis<a class="headerlink" href="#reasoning-analysis" title="Permanent link">¶</a></h3>
|
||
<p><em>Complex problem solving, research, multi-step tasks</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>claude-opus-4-7</code></td>
|
||
<td>Highest Claude capability; temp=0.7.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Cheaper</strong></td>
|
||
<td><code>gemini-2.5-pro</code></td>
|
||
<td>Strong reasoning at 888 ms, $1.25/$10.00; good for research tasks.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Local reasoning</strong></td>
|
||
<td><code>deepseek-r1-distill-llama-70b</code></td>
|
||
<td>R1 chain-of-thought via SAIA at 325 ms; outputs <code><think></code> tokens.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Fast reasoning</strong></td>
|
||
<td><code>cerebras-qwen-3-235b</code></td>
|
||
<td>235B model at 199 ms on Cerebras wafer silicon.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-10-1" name="__codelineno-10-1" href="#__codelineno-10-1"></a>model: claude-opus-4-7 # deep analysis
|
||
<a id="__codelineno-10-2" name="__codelineno-10-2" href="#__codelineno-10-2"></a>model: deepseek-r1-distill-llama-70b # local reasoning
|
||
</code></pre></div>
|
||
<hr />
|
||
<h3 id="batch-offline-processing">Batch / Offline Processing<a class="headerlink" href="#batch-offline-processing" title="Permanent link">¶</a></h3>
|
||
<p><em>Non-real-time document processing — cost-optimised, high throughput</em></p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Role</th>
|
||
<th>Model</th>
|
||
<th>Reasoning</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><strong>Primary</strong></td>
|
||
<td><code>gemini-2.5-flash-lite</code></td>
|
||
<td>$0.10/$0.40; cheapest API model with tools + vision.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Free</strong></td>
|
||
<td><code>llama-3.3-70b-instruct</code></td>
|
||
<td>SAIA self-hosted 70B at 168 ms; 131K context, no API cost.</td>
|
||
</tr>
|
||
<tr>
|
||
<td><strong>Alt</strong></td>
|
||
<td><code>qwen3-30b-a3b-instruct-2507</code></td>
|
||
<td>262K context MoE; good for long-document batch on SAIA.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="highlight"><pre><span></span><code><a id="__codelineno-11-1" name="__codelineno-11-1" href="#__codelineno-11-1"></a>model: gemini-2.5-flash-lite # cost-sensitive API batch
|
||
<a id="__codelineno-11-2" name="__codelineno-11-2" href="#__codelineno-11-2"></a>model: llama-3.3-70b-instruct # free local batch
|
||
</code></pre></div>
|
||
<hr />
|
||
<h2 id="aliases-duplicates">Aliases & Duplicates<a class="headerlink" href="#aliases-duplicates" title="Permanent link">¶</a></h2>
|
||
<p>The following model IDs are aliases that route to the same backend model. Use the canonical ID in production to avoid ambiguity:</p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Alias ID</th>
|
||
<th>Canonical Model</th>
|
||
<th>Notes</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><code>sonnet</code></td>
|
||
<td><code>claude-sonnet-4-6</code></td>
|
||
<td>Short alias</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>opus</code></td>
|
||
<td><code>claude-opus-4-7</code></td>
|
||
<td>Short alias</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>haiku</code></td>
|
||
<td><code>claude-haiku-4-5</code></td>
|
||
<td>Short alias</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>text-embedding-ada-002</code></td>
|
||
<td><code>gemini-embedding-2</code></td>
|
||
<td>OpenAI compat alias</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>text-embedding-3-small</code></td>
|
||
<td><code>gemini-embedding-2</code></td>
|
||
<td>OpenAI compat alias</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>text-embedding-3-large</code></td>
|
||
<td><code>gemini-embedding-2</code></td>
|
||
<td>OpenAI compat alias</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<hr />
|
||
<h2 id="experimental-not-yet-validated">Experimental / Not Yet Validated<a class="headerlink" href="#experimental-not-yet-validated" title="Permanent link">¶</a></h2>
|
||
<p>The following models returned errors or have unresolved issues in live testing:</p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Model ID</th>
|
||
<th>Status</th>
|
||
<th>Error Detail</th>
|
||
<th>Action</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td><code>voxtral-mini-latest</code></td>
|
||
<td>✅ Fixed 2026-05-23</td>
|
||
<td>Stray <code>api_base: saia-image-proxy:5999</code> — deleted + re-added clean; now routes to Mistral API</td>
|
||
<td>—</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>voxtral-mini-realtime-latest</code></td>
|
||
<td>🗑️ Removed 2026-05-23</td>
|
||
<td>WebSocket-only realtime endpoint; incompatible with REST completions</td>
|
||
<td>Removed from LiteLLM; use Mistral WS API directly if needed</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>mistral-ocr-latest</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>OCR-mode model; requires document input, not chat completions</td>
|
||
<td>Use via dedicated OCR pipeline only</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>voxtral-mini-transcribe-2507</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Audio transcription; not a chat completions model</td>
|
||
<td>Use via audio transcription endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>gemini-2.5-flash-image</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Image generation; not a chat completions model</td>
|
||
<td>Use via images endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>saia-flux</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>FLUX image generation</td>
|
||
<td>Use via images endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>saia-image-edit</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Image editing</td>
|
||
<td>Use via image edit endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>tts-1-de</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Piper TTS audio output</td>
|
||
<td>Use via audio/speech endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>tts-1</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Kokoro-82M TTS</td>
|
||
<td>Use via audio/speech endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>voxtral-mini-tts-latest</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Mistral TTS</td>
|
||
<td>Use via audio/speech endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>saia-whisper</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Whisper ASR</td>
|
||
<td>Use via audio/transcriptions endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>whisper-1</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>faster-whisper ASR</td>
|
||
<td>Use via audio/transcriptions endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>whisper-large-v3-turbo</code></td>
|
||
<td>⚠️ Not benchmarked</td>
|
||
<td>Groq Whisper ASR</td>
|
||
<td>Use via audio/transcriptions endpoint</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>gemini-3.1-flash-lite</code></td>
|
||
<td>⚠️ Context unknown</td>
|
||
<td>Preview model; ctx window not documented</td>
|
||
<td>Monitor Gemini API release notes</td>
|
||
</tr>
|
||
<tr>
|
||
<td><code>cerebras-qwen-3-235b</code></td>
|
||
<td>⚠️ Context unknown</td>
|
||
<td>Context window not documented in LiteLLM config</td>
|
||
<td>Check Cerebras API docs</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<hr />
|
||
<h2 id="raw-benchmark-data">Raw Benchmark Data<a class="headerlink" href="#raw-benchmark-data" title="Permanent link">¶</a></h2>
|
||
<p>All measurements from <code>2026-05-23</code>. Single <code>max_tokens=5</code> completion, prompt: <em>"Reply with exactly: ok"</em>.</p>
|
||
<table>
|
||
<thead>
|
||
<tr>
|
||
<th>Model ID</th>
|
||
<th>HTTP Status</th>
|
||
<th>Latency (ms)</th>
|
||
<th>Prompt Tokens</th>
|
||
<th>Completion Tokens</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td>claude-sonnet-4-6</td>
|
||
<td>200</td>
|
||
<td>2 083</td>
|
||
<td>3</td>
|
||
<td>4</td>
|
||
</tr>
|
||
<tr>
|
||
<td>claude-opus-4-7</td>
|
||
<td>200</td>
|
||
<td>3 218</td>
|
||
<td>6</td>
|
||
<td>6</td>
|
||
</tr>
|
||
<tr>
|
||
<td>claude-haiku-4-5</td>
|
||
<td>200</td>
|
||
<td>1 453</td>
|
||
<td>10</td>
|
||
<td>41</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-small-latest</td>
|
||
<td>200</td>
|
||
<td>160</td>
|
||
<td>8</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>mistral-small-latest</td>
|
||
<td>200</td>
|
||
<td>222</td>
|
||
<td>20</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-mini-latest</td>
|
||
<td>500</td>
|
||
<td>48</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
</tr>
|
||
<tr>
|
||
<td>codestral-latest</td>
|
||
<td>200</td>
|
||
<td>275</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>devstral-2-123b-instruct-2512</td>
|
||
<td>200</td>
|
||
<td>157</td>
|
||
<td>8</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3-coder-30b-a3b-instruct</td>
|
||
<td>200</td>
|
||
<td>5</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>openai-gpt-oss-120b</td>
|
||
<td>200</td>
|
||
<td>118</td>
|
||
<td>74</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-flash</td>
|
||
<td>200</td>
|
||
<td>501</td>
|
||
<td>6</td>
|
||
<td>1</td>
|
||
</tr>
|
||
<tr>
|
||
<td>cerebras-llama-3.1-8b</td>
|
||
<td>200</td>
|
||
<td>258</td>
|
||
<td>40</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>meta-llama/llama-4-scout-17b-16e-instruct</td>
|
||
<td>200</td>
|
||
<td>442</td>
|
||
<td>15</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>mistral-large-latest</td>
|
||
<td>200</td>
|
||
<td>309</td>
|
||
<td>8</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-flash-lite</td>
|
||
<td>200</td>
|
||
<td>518</td>
|
||
<td>6</td>
|
||
<td>1</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3-omni-30b-a3b-instruct</td>
|
||
<td>200</td>
|
||
<td>178</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>llama-3.3-70b-instruct</td>
|
||
<td>200</td>
|
||
<td>168</td>
|
||
<td>102</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>deepseek-r1-distill-llama-70b</td>
|
||
<td>200</td>
|
||
<td>325</td>
|
||
<td>8</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-2.5-pro</td>
|
||
<td>200</td>
|
||
<td>888</td>
|
||
<td>6</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>pixtral-large-latest</td>
|
||
<td>200</td>
|
||
<td>27</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>voxtral-mini-realtime-latest</td>
|
||
<td>400</td>
|
||
<td>140</td>
|
||
<td>—</td>
|
||
<td>—</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-35b-a3b</td>
|
||
<td>200</td>
|
||
<td>250</td>
|
||
<td>15</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-27b</td>
|
||
<td>200</td>
|
||
<td>266</td>
|
||
<td>15</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.6-35b-a3b</td>
|
||
<td>200</td>
|
||
<td>204</td>
|
||
<td>15</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>apertus-70b-instruct-2509</td>
|
||
<td>200</td>
|
||
<td>143</td>
|
||
<td>66</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>glm-4.7</td>
|
||
<td>200</td>
|
||
<td>210</td>
|
||
<td>9</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-3.1-pro-preview</td>
|
||
<td>200</td>
|
||
<td>1 215</td>
|
||
<td>6</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-3-pro-preview</td>
|
||
<td>200</td>
|
||
<td>1 258</td>
|
||
<td>6</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3-30b-a3b-instruct-2507</td>
|
||
<td>200</td>
|
||
<td>127</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemini-3.1-flash-lite</td>
|
||
<td>200</td>
|
||
<td>482</td>
|
||
<td>6</td>
|
||
<td>1</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemma-3-27b-it</td>
|
||
<td>200</td>
|
||
<td>284</td>
|
||
<td>14</td>
|
||
<td>3</td>
|
||
</tr>
|
||
<tr>
|
||
<td>internvl3.5-30b-a3b</td>
|
||
<td>200</td>
|
||
<td>116</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-122b-a10b</td>
|
||
<td>200</td>
|
||
<td>372</td>
|
||
<td>15</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>qwen3.5-397b-a17b</td>
|
||
<td>200</td>
|
||
<td>248</td>
|
||
<td>15</td>
|
||
<td>5</td>
|
||
</tr>
|
||
<tr>
|
||
<td>gemma-4-31b-it</td>
|
||
<td>200</td>
|
||
<td>177</td>
|
||
<td>18</td>
|
||
<td>2</td>
|
||
</tr>
|
||
<tr>
|
||
<td>cerebras-qwen-3-235b</td>
|
||
<td>200</td>
|
||
<td>199</td>
|
||
<td>13</td>
|
||
<td>2</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<blockquote>
|
||
<p><strong>Benchmark note on SAIA models with < 50 ms latency</strong> (qwen3-coder: 5 ms, pixtral: 27 ms): these figures reflect a KV-cache or pre-warmed response for the trivial prompt. Real-world TTFT for cold prompts will be 100–400 ms depending on model size and GPU availability.</p>
|
||
</blockquote>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
</article>
|
||
</div>
|
||
|
||
|
||
<script>var target=document.getElementById(location.hash.slice(1));target&&target.name&&(target.checked=target.name.startsWith("__tabbed_"))</script>
|
||
</div>
|
||
|
||
<button type="button" class="md-top md-icon" data-md-component="top" hidden>
|
||
|
||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M13 20h-2V8l-5.5 5.5-1.42-1.42L12 4.16l7.92 7.92-1.42 1.42L13 8z"/></svg>
|
||
Back to top
|
||
</button>
|
||
|
||
</main>
|
||
|
||
<footer class="md-footer">
|
||
|
||
<div class="md-footer-meta md-typeset">
|
||
<div class="md-footer-meta__inner md-grid">
|
||
<div class="md-copyright">
|
||
|
||
|
||
Made with
|
||
<a href="https://squidfunk.github.io/mkdocs-material/" target="_blank" rel="noopener">
|
||
Material for MkDocs
|
||
</a>
|
||
|
||
</div>
|
||
|
||
</div>
|
||
</div>
|
||
</footer>
|
||
|
||
</div>
|
||
<div class="md-dialog" data-md-component="dialog">
|
||
<div class="md-dialog__inner md-typeset"></div>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
|
||
<script id="__config" type="application/json">{"annotate": null, "base": "../..", "features": ["navigation.tabs", "navigation.sections", "navigation.expand", "navigation.top", "search.highlight", "search.suggest", "content.code.copy"], "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}</script>
|
||
|
||
|
||
<script src="../../assets/javascripts/bundle.79ae519e.min.js"></script>
|
||
|
||
|
||
</body>
|
||
</html> |