You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

index.html 35 kB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980818283848586878889909192939495969798991001011021031041051061071081091101111121131141151161171181191201211221231241251261271281291301311321331341351361371381391401411421431441451461471481491501511521531541551561571581591601611621631641651661671681691701711721731741751761771781791801811821831841851861871881891901911921931941951961971981992002012022032042052062072082092102112122132142152162172182192202212222232242252262272282292302312322332342352362372382392402412422432442452462472482492502512522532542552562572582592602612622632642652662672682692702712722732742752762772782792802812822832842852862872882892902912922932942952962972982993003013023033043053063073083093103113123133143153163173183193203213223233243253263273283293303313323333343353363373383393403413423433443453463473483493503513523533543553563573583593603613623633643653663673683693703713723733743753763773783793803813823833843853863873883893903913923933943953963973983994004014024034044054064074084094104114124134144154164174184194204214224234244254264274284294304314324334344354364374384394404414424434444454464474484494504514524534544554564574584594604614624634644654664674684694704714724734744754764774784794804814824834844854864874884894904914924934944954964974984995005015025035045055065075085095105115125135145155165175185195205215225235245255265275285295305315325335345355365375385395405415425435445455465475485495505515525535545555565575585595605615625635645655665675685695705715725735745755765775785795805815825835845855865875885895905915925935945955965975985996006016026036046056066076086096106116126136146156166176186196206216226236246256266276286296306316326336346356366376386396406416426436446456466476486496506516526536546556566576586596606616626636646656666676686696706716726736746756766776786796806816826836846856866876886896906916926936946956966976986997007017027037047057067077087097107117127137147157167177187197207217227237247257267277287297307317327337347357367377387397407417427437447457467477487497507517527537547557567577587597607617627637647657667677687697707717727737747757767777787797807817827837847857867877887897907917927937947957967977987998008018028038048058068078088098108118128138148158168178188198208218228238248258268278288298308318328338348358368378388398408418428438448458468478488498508518528538548558568578588598608618628638648658668678688698708718728738748758768778788798808818828838848858868878888898908918928938948958968978988999009019029039049059069079089099109119129139149159169179189199209219229239249259269279289299309319329339349359369379389399409419429439449459469479489499509519529539549559569579589599609619629639649659669679689699709719729739749759769779789799809819829839849859869879889899909919929939949959969979989991000100110021003100410051006100710081009101010111012101310141015101610171018101910201021102210231024102510261027102810291030103110321033103410351036103710381039104010411042104310441045104610471048104910501051105210531054105510561057105810591060106110621063106410651066106710681069107010711072107310741075107610771078107910801081108210831084108510861087108810891090109110921093109410951096109710981099110011011102110311041105110611071108110911101111111211131114111511161117111811191120112111221123112411251126112711281129113011311132113311341135113611371138113911401141114211431144114511461147114811491150115111521153115411551156115711581159116011611162116311641165116611671168116911701171117211731174117511761177117811791180118111821183118411851186118711881189119011911192119311941195119611971198119912001201120212031204120512061207120812091210121112121213121412151216121712181219122012211222122312241225122612271228122912301231123212331234123512361237123812391240124112421243124412451246124712481249125012511252125312541255125612571258125912601261126212631264126512661267126812691270127112721273127412751276127712781279128012811282128312841285128612871288128912901291129212931294129512961297129812991300130113021303130413051306130713081309131013111312131313141315131613171318131913201321132213231324132513261327132813291330133113321333133413351336133713381339134013411342134313441345134613471348134913501351135213531354135513561357135813591360136113621363136413651366136713681369137013711372137313741375137613771378137913801381138213831384138513861387138813891390139113921393139413951396139713981399140014011402140314041405140614071408140914101411141214131414141514161417141814191420142114221423142414251426142714281429143014311432143314341435143614371438143914401441144214431444144514461447144814491450145114521453145414551456145714581459146014611462146314641465146614671468146914701471147214731474147514761477147814791480148114821483148414851486148714881489149014911492149314941495149614971498149915001501150215031504150515061507150815091510151115121513151415151516151715181519152015211522152315241525152615271528152915301531153215331534153515361537153815391540154115421543154415451546154715481549155015511552155315541555155615571558155915601561156215631564156515661567156815691570157115721573157415751576157715781579158015811582158315841585158615871588158915901591159215931594159515961597159815991600160116021603160416051606160716081609161016111612161316141615161616171618161916201621162216231624162516261627162816291630163116321633163416351636163716381639164016411642164316441645164616471648164916501651165216531654165516561657165816591660166116621663166416651666166716681669167016711672167316741675167616771678167916801681168216831684168516861687168816891690169116921693169416951696169716981699170017011702170317041705170617071708170917101711171217131714171517161717171817191720172117221723172417251726172717281729173017311732173317341735
  1. <!doctype html>
  2. <html lang="en" class="no-js">
  3. <head>
  4. <meta charset="utf-8">
  5. <meta name="viewport" content="width=device-width,initial-scale=1">
  6. <link rel="prev" href="../embeddings/">
  7. <link rel="next" href="../save-load-state/">
  8. <link rel="icon" href="../../assets/images/favicon.png">
  9. <meta name="generator" content="mkdocs-1.4.3, mkdocs-material-9.1.16">
  10. <title>Quantization - LLamaSharp Documentation</title>
  11. <link rel="stylesheet" href="../../assets/stylesheets/main.26e3688c.min.css">
  12. <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  13. <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">
  14. <style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
  15. <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>
  16. </head>
  17. <body dir="ltr">
  18. <script>var palette=__md_get("__palette");if(palette&&"object"==typeof palette.color)for(var key of Object.keys(palette.color))document.body.setAttribute("data-md-color-"+key,palette.color[key])</script>
  19. <input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
  20. <input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
  21. <label class="md-overlay" for="__drawer"></label>
  22. <div data-md-component="skip">
  23. <a href="#quantization" class="md-skip">
  24. Skip to content
  25. </a>
  26. </div>
  27. <div data-md-component="announce">
  28. </div>
  29. <div data-md-color-scheme="default" data-md-component="outdated" hidden>
  30. </div>
  31. <header class="md-header md-header--shadow" data-md-component="header">
  32. <nav class="md-header__inner md-grid" aria-label="Header">
  33. <a href="../.." title="LLamaSharp Documentation" class="md-header__button md-logo" aria-label="LLamaSharp Documentation" data-md-component="logo">
  34. <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 3.54Z"/></svg>
  35. </a>
  36. <label class="md-header__button md-icon" for="__drawer">
  37. <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3V6m0 5h18v2H3v-2m0 5h18v2H3v-2Z"/></svg>
  38. </label>
  39. <div class="md-header__title" data-md-component="header-title">
  40. <div class="md-header__ellipsis">
  41. <div class="md-header__topic">
  42. <span class="md-ellipsis">
  43. LLamaSharp Documentation
  44. </span>
  45. </div>
  46. <div class="md-header__topic" data-md-component="header-topic">
  47. <span class="md-ellipsis">
  48. Quantization
  49. </span>
  50. </div>
  51. </div>
  52. </div>
  53. <label class="md-header__button md-icon" for="__search">
  54. <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-.27A6.516 6.516 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 5Z"/></svg>
  55. </label>
  56. <div class="md-search" data-md-component="search" role="dialog">
  57. <label class="md-search__overlay" for="__search"></label>
  58. <div class="md-search__inner" role="search">
  59. <form class="md-search__form" name="search">
  60. <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>
  61. <label class="md-search__icon md-icon" for="__search">
  62. <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-.27A6.516 6.516 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 5Z"/></svg>
  63. <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 11h12Z"/></svg>
  64. </label>
  65. <nav class="md-search__options" aria-label="Search">
  66. <button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
  67. <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 12 19 6.41Z"/></svg>
  68. </button>
  69. </nav>
  70. </form>
  71. <div class="md-search__output">
  72. <div class="md-search__scrollwrap" data-md-scrollfix>
  73. <div class="md-search-result" data-md-component="search-result">
  74. <div class="md-search-result__meta">
  75. Initializing search
  76. </div>
  77. <ol class="md-search-result__list" role="presentation"></ol>
  78. </div>
  79. </div>
  80. </div>
  81. </div>
  82. </div>
  83. </nav>
  84. </header>
  85. <div class="md-container" data-md-component="container">
  86. <main class="md-main" data-md-component="main">
  87. <div class="md-main__inner md-grid">
  88. <div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
  89. <div class="md-sidebar__scrollwrap">
  90. <div class="md-sidebar__inner">
  91. <nav class="md-nav md-nav--primary" aria-label="Navigation" data-md-level="0">
  92. <label class="md-nav__title" for="__drawer">
  93. <a href="../.." title="LLamaSharp Documentation" class="md-nav__button md-logo" aria-label="LLamaSharp Documentation" data-md-component="logo">
  94. <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 3.54Z"/></svg>
  95. </a>
  96. LLamaSharp Documentation
  97. </label>
  98. <ul class="md-nav__list" data-md-scrollfix>
  99. <li class="md-nav__item">
  100. <a href="../.." class="md-nav__link">
  101. Overview
  102. </a>
  103. </li>
  104. <li class="md-nav__item">
  105. <a href="../../GetStarted/" class="md-nav__link">
  106. Get Started
  107. </a>
  108. </li>
  109. <li class="md-nav__item">
  110. <a href="../../Architecher/" class="md-nav__link">
  111. Architecher
  112. </a>
  113. </li>
  114. <li class="md-nav__item">
  115. <a href="../../Tricks/" class="md-nav__link">
  116. Tricks for FAQ
  117. </a>
  118. </li>
  119. <li class="md-nav__item">
  120. <a href="../../ContributingGuide/" class="md-nav__link">
  121. Contributing Guide
  122. </a>
  123. </li>
  124. <li class="md-nav__item md-nav__item--active md-nav__item--nested">
  125. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_6" checked>
  126. <label class="md-nav__link" for="__nav_6" id="__nav_6_label" tabindex="0">
  127. LLamaModel
  128. <span class="md-nav__icon md-icon"></span>
  129. </label>
  130. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_6_label" aria-expanded="true">
  131. <label class="md-nav__title" for="__nav_6">
  132. <span class="md-nav__icon md-icon"></span>
  133. LLamaModel
  134. </label>
  135. <ul class="md-nav__list" data-md-scrollfix>
  136. <li class="md-nav__item">
  137. <a href="../parameters/" class="md-nav__link">
  138. Model Parameters
  139. </a>
  140. </li>
  141. <li class="md-nav__item">
  142. <a href="../tokenization/" class="md-nav__link">
  143. Tokenization
  144. </a>
  145. </li>
  146. <li class="md-nav__item">
  147. <a href="../embeddings/" class="md-nav__link">
  148. Get Embeddings
  149. </a>
  150. </li>
  151. <li class="md-nav__item md-nav__item--active">
  152. <input class="md-nav__toggle md-toggle" type="checkbox" id="__toc">
  153. <a href="./" class="md-nav__link md-nav__link--active">
  154. Quantization
  155. </a>
  156. </li>
  157. <li class="md-nav__item">
  158. <a href="../save-load-state/" class="md-nav__link">
  159. Save/Load State
  160. </a>
  161. </li>
  162. </ul>
  163. </nav>
  164. </li>
  165. <li class="md-nav__item md-nav__item--nested">
  166. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_7" >
  167. <label class="md-nav__link" for="__nav_7" id="__nav_7_label" tabindex="0">
  168. LLamaExecutors
  169. <span class="md-nav__icon md-icon"></span>
  170. </label>
  171. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_7_label" aria-expanded="false">
  172. <label class="md-nav__title" for="__nav_7">
  173. <span class="md-nav__icon md-icon"></span>
  174. LLamaExecutors
  175. </label>
  176. <ul class="md-nav__list" data-md-scrollfix>
  177. <li class="md-nav__item">
  178. <a href="../../LLamaExecutors/parameters/" class="md-nav__link">
  179. Inference Parameters
  180. </a>
  181. </li>
  182. <li class="md-nav__item">
  183. <a href="../../LLamaExecutors/text-to-text-apis/" class="md-nav__link">
  184. Text-to-Text APIs
  185. </a>
  186. </li>
  187. <li class="md-nav__item">
  188. <a href="../../LLamaExecutors/save-load-state/" class="md-nav__link">
  189. Save/Load State
  190. </a>
  191. </li>
  192. <li class="md-nav__item">
  193. <a href="../../LLamaExecutors/differences/" class="md-nav__link">
  194. Differences of Executors
  195. </a>
  196. </li>
  197. </ul>
  198. </nav>
  199. </li>
  200. <li class="md-nav__item md-nav__item--nested">
  201. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_8" >
  202. <label class="md-nav__link" for="__nav_8" id="__nav_8_label" tabindex="0">
  203. ChatSession
  204. <span class="md-nav__icon md-icon"></span>
  205. </label>
  206. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_8_label" aria-expanded="false">
  207. <label class="md-nav__title" for="__nav_8">
  208. <span class="md-nav__icon md-icon"></span>
  209. ChatSession
  210. </label>
  211. <ul class="md-nav__list" data-md-scrollfix>
  212. <li class="md-nav__item">
  213. <a href="../../ChatSession/basic-usages/" class="md-nav__link">
  214. Basic Usages
  215. </a>
  216. </li>
  217. <li class="md-nav__item">
  218. <a href="../../ChatSession/transforms/" class="md-nav__link">
  219. Transoforms
  220. </a>
  221. </li>
  222. <li class="md-nav__item">
  223. <a href="../../ChatSession/save-load-session/" class="md-nav__link">
  224. Save/Load Session
  225. </a>
  226. </li>
  227. </ul>
  228. </nav>
  229. </li>
  230. <li class="md-nav__item md-nav__item--nested">
  231. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_9" >
  232. <label class="md-nav__link" for="__nav_9" id="__nav_9_label" tabindex="0">
  233. Non-English Usages
  234. <span class="md-nav__icon md-icon"></span>
  235. </label>
  236. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_9_label" aria-expanded="false">
  237. <label class="md-nav__title" for="__nav_9">
  238. <span class="md-nav__icon md-icon"></span>
  239. Non-English Usages
  240. </label>
  241. <ul class="md-nav__list" data-md-scrollfix>
  242. <li class="md-nav__item">
  243. <a href="../../NonEnglishUsage/Chinese/" class="md-nav__link">
  244. Chinese
  245. </a>
  246. </li>
  247. </ul>
  248. </nav>
  249. </li>
  250. <li class="md-nav__item md-nav__item--nested">
  251. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_10" >
  252. <label class="md-nav__link" for="__nav_10" id="__nav_10_label" tabindex="0">
  253. High-level Applications
  254. <span class="md-nav__icon md-icon"></span>
  255. </label>
  256. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_10_label" aria-expanded="false">
  257. <label class="md-nav__title" for="__nav_10">
  258. <span class="md-nav__icon md-icon"></span>
  259. High-level Applications
  260. </label>
  261. <ul class="md-nav__list" data-md-scrollfix>
  262. <li class="md-nav__item">
  263. <a href="../../HighLevelApps/bot-sharp/" class="md-nav__link">
  264. BotSharp
  265. </a>
  266. </li>
  267. </ul>
  268. </nav>
  269. </li>
  270. <li class="md-nav__item md-nav__item--nested">
  271. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_11" >
  272. <label class="md-nav__link" for="__nav_11" id="__nav_11_label" tabindex="0">
  273. More
  274. <span class="md-nav__icon md-icon"></span>
  275. </label>
  276. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_11_label" aria-expanded="false">
  277. <label class="md-nav__title" for="__nav_11">
  278. <span class="md-nav__icon md-icon"></span>
  279. More
  280. </label>
  281. <ul class="md-nav__list" data-md-scrollfix>
  282. <li class="md-nav__item">
  283. <a href="../../More/log/" class="md-nav__link">
  284. Logger
  285. </a>
  286. </li>
  287. </ul>
  288. </nav>
  289. </li>
  290. <li class="md-nav__item md-nav__item--nested">
  291. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_12" >
  292. <label class="md-nav__link" for="__nav_12" id="__nav_12_label" tabindex="0">
  293. Examples
  294. <span class="md-nav__icon md-icon"></span>
  295. </label>
  296. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_12_label" aria-expanded="false">
  297. <label class="md-nav__title" for="__nav_12">
  298. <span class="md-nav__icon md-icon"></span>
  299. Examples
  300. </label>
  301. <ul class="md-nav__list" data-md-scrollfix>
  302. <li class="md-nav__item">
  303. <a href="../../Examples/ChatSessionStripRoleName/" class="md-nav__link">
  304. Chat session 1
  305. </a>
  306. </li>
  307. <li class="md-nav__item">
  308. <a href="../../Examples/ChatSessionWithRoleName/" class="md-nav__link">
  309. Chat session 2
  310. </a>
  311. </li>
  312. <li class="md-nav__item">
  313. <a href="../../Examples/GetEmbeddings/" class="md-nav__link">
  314. Get embeddings
  315. </a>
  316. </li>
  317. <li class="md-nav__item">
  318. <a href="../../Examples/InstructModeExecute/" class="md-nav__link">
  319. Instruct executor
  320. </a>
  321. </li>
  322. <li class="md-nav__item">
  323. <a href="../../Examples/InteractiveModeExecute/" class="md-nav__link">
  324. Interactive executor
  325. </a>
  326. </li>
  327. <li class="md-nav__item">
  328. <a href="../../Examples/StatelessModeExecute/" class="md-nav__link">
  329. Stateless exeutor
  330. </a>
  331. </li>
  332. <li class="md-nav__item">
  333. <a href="../../Examples/LoadAndSaveSession/" class="md-nav__link">
  334. Load/Save session
  335. </a>
  336. </li>
  337. <li class="md-nav__item">
  338. <a href="../../Examples/LoadAndSaveState/" class="md-nav__link">
  339. Load/Save state
  340. </a>
  341. </li>
  342. <li class="md-nav__item">
  343. <a href="../../Examples/QuantizeModel/" class="md-nav__link">
  344. Quantize model
  345. </a>
  346. </li>
  347. </ul>
  348. </nav>
  349. </li>
  350. <li class="md-nav__item md-nav__item--nested">
  351. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_13" >
  352. <label class="md-nav__link" for="__nav_13" id="__nav_13_label" tabindex="0">
  353. API Reference
  354. <span class="md-nav__icon md-icon"></span>
  355. </label>
  356. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_13_label" aria-expanded="false">
  357. <label class="md-nav__title" for="__nav_13">
  358. <span class="md-nav__icon md-icon"></span>
  359. API Reference
  360. </label>
  361. <ul class="md-nav__list" data-md-scrollfix>
  362. <li class="md-nav__item">
  363. <a href="../../xmldocs/" class="md-nav__link">
  364. index
  365. </a>
  366. </li>
  367. <li class="md-nav__item">
  368. <a href="../../xmldocs/llama.abstractions.ihistorytransform/" class="md-nav__link">
  369. llama.abstractions.ihistorytransform
  370. </a>
  371. </li>
  372. <li class="md-nav__item">
  373. <a href="../../xmldocs/llama.abstractions.illamaexecutor/" class="md-nav__link">
  374. llama.abstractions.illamaexecutor
  375. </a>
  376. </li>
  377. <li class="md-nav__item">
  378. <a href="../../xmldocs/llama.abstractions.itextstreamtransform/" class="md-nav__link">
  379. llama.abstractions.itextstreamtransform
  380. </a>
  381. </li>
  382. <li class="md-nav__item">
  383. <a href="../../xmldocs/llama.abstractions.itexttransform/" class="md-nav__link">
  384. llama.abstractions.itexttransform
  385. </a>
  386. </li>
  387. <li class="md-nav__item">
  388. <a href="../../xmldocs/llama.chatsession/" class="md-nav__link">
  389. llama.chatsession
  390. </a>
  391. </li>
  392. <li class="md-nav__item">
  393. <a href="../../xmldocs/llama.common.authorrole/" class="md-nav__link">
  394. llama.common.authorrole
  395. </a>
  396. </li>
  397. <li class="md-nav__item">
  398. <a href="../../xmldocs/llama.common.chathistory/" class="md-nav__link">
  399. llama.common.chathistory
  400. </a>
  401. </li>
  402. <li class="md-nav__item">
  403. <a href="../../xmldocs/llama.common.fixedsizequeue-1/" class="md-nav__link">
  404. llama.common.fixedsizequeue-1
  405. </a>
  406. </li>
  407. <li class="md-nav__item">
  408. <a href="../../xmldocs/llama.common.illamalogger/" class="md-nav__link">
  409. llama.common.illamalogger
  410. </a>
  411. </li>
  412. <li class="md-nav__item">
  413. <a href="../../xmldocs/llama.common.inferenceparams/" class="md-nav__link">
  414. llama.common.inferenceparams
  415. </a>
  416. </li>
  417. <li class="md-nav__item">
  418. <a href="../../xmldocs/llama.common.llamadefaultlogger/" class="md-nav__link">
  419. llama.common.llamadefaultlogger
  420. </a>
  421. </li>
  422. <li class="md-nav__item">
  423. <a href="../../xmldocs/llama.common.mirostatetype/" class="md-nav__link">
  424. llama.common.mirostatetype
  425. </a>
  426. </li>
  427. <li class="md-nav__item">
  428. <a href="../../xmldocs/llama.common.modelparams/" class="md-nav__link">
  429. llama.common.modelparams
  430. </a>
  431. </li>
  432. <li class="md-nav__item">
  433. <a href="../../xmldocs/llama.exceptions.runtimeerror/" class="md-nav__link">
  434. llama.exceptions.runtimeerror
  435. </a>
  436. </li>
  437. <li class="md-nav__item">
  438. <a href="../../xmldocs/llama.extensions.dictionaryextension/" class="md-nav__link">
  439. llama.extensions.dictionaryextension
  440. </a>
  441. </li>
  442. <li class="md-nav__item">
  443. <a href="../../xmldocs/llama.instructexecutor/" class="md-nav__link">
  444. llama.instructexecutor
  445. </a>
  446. </li>
  447. <li class="md-nav__item">
  448. <a href="../../xmldocs/llama.interactiveexecutor/" class="md-nav__link">
  449. llama.interactiveexecutor
  450. </a>
  451. </li>
  452. <li class="md-nav__item">
  453. <a href="../../xmldocs/llama.llamaembedder/" class="md-nav__link">
  454. llama.llamaembedder
  455. </a>
  456. </li>
  457. <li class="md-nav__item">
  458. <a href="../../xmldocs/llama.llamamodel/" class="md-nav__link">
  459. llama.llamamodel
  460. </a>
  461. </li>
  462. <li class="md-nav__item">
  463. <a href="../../xmldocs/llama.llamaquantizer/" class="md-nav__link">
  464. llama.llamaquantizer
  465. </a>
  466. </li>
  467. <li class="md-nav__item">
  468. <a href="../../xmldocs/llama.llamatransforms/" class="md-nav__link">
  469. llama.llamatransforms
  470. </a>
  471. </li>
  472. <li class="md-nav__item">
  473. <a href="../../xmldocs/llama.native.llamacontextparams/" class="md-nav__link">
  474. llama.native.llamacontextparams
  475. </a>
  476. </li>
  477. <li class="md-nav__item">
  478. <a href="../../xmldocs/llama.native.llamaftype/" class="md-nav__link">
  479. llama.native.llamaftype
  480. </a>
  481. </li>
  482. <li class="md-nav__item">
  483. <a href="../../xmldocs/llama.native.llamatokendata/" class="md-nav__link">
  484. llama.native.llamatokendata
  485. </a>
  486. </li>
  487. <li class="md-nav__item">
  488. <a href="../../xmldocs/llama.native.llamatokendataarray/" class="md-nav__link">
  489. llama.native.llamatokendataarray
  490. </a>
  491. </li>
  492. <li class="md-nav__item">
  493. <a href="../../xmldocs/llama.native.llamatokendataarraynative/" class="md-nav__link">
  494. llama.native.llamatokendataarraynative
  495. </a>
  496. </li>
  497. <li class="md-nav__item">
  498. <a href="../../xmldocs/llama.native.nativeapi/" class="md-nav__link">
  499. llama.native.nativeapi
  500. </a>
  501. </li>
  502. <li class="md-nav__item">
  503. <a href="../../xmldocs/llama.native.safellamacontexthandle/" class="md-nav__link">
  504. llama.native.safellamacontexthandle
  505. </a>
  506. </li>
  507. <li class="md-nav__item">
  508. <a href="../../xmldocs/llama.native.safellamahandlebase/" class="md-nav__link">
  509. llama.native.safellamahandlebase
  510. </a>
  511. </li>
  512. <li class="md-nav__item">
  513. <a href="../../xmldocs/llama.oldversion.chatcompletion/" class="md-nav__link">
  514. llama.oldversion.chatcompletion
  515. </a>
  516. </li>
  517. <li class="md-nav__item">
  518. <a href="../../xmldocs/llama.oldversion.chatcompletionchoice/" class="md-nav__link">
  519. llama.oldversion.chatcompletionchoice
  520. </a>
  521. </li>
  522. <li class="md-nav__item">
  523. <a href="../../xmldocs/llama.oldversion.chatcompletionchunk/" class="md-nav__link">
  524. llama.oldversion.chatcompletionchunk
  525. </a>
  526. </li>
  527. <li class="md-nav__item">
  528. <a href="../../xmldocs/llama.oldversion.chatcompletionchunkchoice/" class="md-nav__link">
  529. llama.oldversion.chatcompletionchunkchoice
  530. </a>
  531. </li>
  532. <li class="md-nav__item">
  533. <a href="../../xmldocs/llama.oldversion.chatcompletionchunkdelta/" class="md-nav__link">
  534. llama.oldversion.chatcompletionchunkdelta
  535. </a>
  536. </li>
  537. <li class="md-nav__item">
  538. <a href="../../xmldocs/llama.oldversion.chatcompletionmessage/" class="md-nav__link">
  539. llama.oldversion.chatcompletionmessage
  540. </a>
  541. </li>
  542. <li class="md-nav__item">
  543. <a href="../../xmldocs/llama.oldversion.chatmessagerecord/" class="md-nav__link">
  544. llama.oldversion.chatmessagerecord
  545. </a>
  546. </li>
  547. <li class="md-nav__item">
  548. <a href="../../xmldocs/llama.oldversion.chatrole/" class="md-nav__link">
  549. llama.oldversion.chatrole
  550. </a>
  551. </li>
  552. <li class="md-nav__item">
  553. <a href="../../xmldocs/llama.oldversion.chatsession-1/" class="md-nav__link">
  554. llama.oldversion.chatsession-1
  555. </a>
  556. </li>
  557. <li class="md-nav__item">
  558. <a href="../../xmldocs/llama.oldversion.completion/" class="md-nav__link">
  559. llama.oldversion.completion
  560. </a>
  561. </li>
  562. <li class="md-nav__item">
  563. <a href="../../xmldocs/llama.oldversion.completionchoice/" class="md-nav__link">
  564. llama.oldversion.completionchoice
  565. </a>
  566. </li>
  567. <li class="md-nav__item">
  568. <a href="../../xmldocs/llama.oldversion.completionchunk/" class="md-nav__link">
  569. llama.oldversion.completionchunk
  570. </a>
  571. </li>
  572. <li class="md-nav__item">
  573. <a href="../../xmldocs/llama.oldversion.completionlogprobs/" class="md-nav__link">
  574. llama.oldversion.completionlogprobs
  575. </a>
  576. </li>
  577. <li class="md-nav__item">
  578. <a href="../../xmldocs/llama.oldversion.completionusage/" class="md-nav__link">
  579. llama.oldversion.completionusage
  580. </a>
  581. </li>
  582. <li class="md-nav__item">
  583. <a href="../../xmldocs/llama.oldversion.embedding/" class="md-nav__link">
  584. llama.oldversion.embedding
  585. </a>
  586. </li>
  587. <li class="md-nav__item">
  588. <a href="../../xmldocs/llama.oldversion.embeddingdata/" class="md-nav__link">
  589. llama.oldversion.embeddingdata
  590. </a>
  591. </li>
  592. <li class="md-nav__item">
  593. <a href="../../xmldocs/llama.oldversion.embeddingusage/" class="md-nav__link">
  594. llama.oldversion.embeddingusage
  595. </a>
  596. </li>
  597. <li class="md-nav__item">
  598. <a href="../../xmldocs/llama.oldversion.ichatmodel/" class="md-nav__link">
  599. llama.oldversion.ichatmodel
  600. </a>
  601. </li>
  602. <li class="md-nav__item">
  603. <a href="../../xmldocs/llama.oldversion.llamaembedder/" class="md-nav__link">
  604. llama.oldversion.llamaembedder
  605. </a>
  606. </li>
  607. <li class="md-nav__item">
  608. <a href="../../xmldocs/llama.oldversion.llamamodel/" class="md-nav__link">
  609. llama.oldversion.llamamodel
  610. </a>
  611. </li>
  612. <li class="md-nav__item">
  613. <a href="../../xmldocs/llama.oldversion.llamaparams/" class="md-nav__link">
  614. llama.oldversion.llamaparams
  615. </a>
  616. </li>
  617. <li class="md-nav__item">
  618. <a href="../../xmldocs/llama.resettablellamamodel/" class="md-nav__link">
  619. llama.resettablellamamodel
  620. </a>
  621. </li>
  622. <li class="md-nav__item">
  623. <a href="../../xmldocs/llama.statefulexecutorbase/" class="md-nav__link">
  624. llama.statefulexecutorbase
  625. </a>
  626. </li>
  627. <li class="md-nav__item">
  628. <a href="../../xmldocs/llama.statelessexecutor/" class="md-nav__link">
  629. llama.statelessexecutor
  630. </a>
  631. </li>
  632. </ul>
  633. </nav>
  634. </li>
  635. </ul>
  636. </nav>
  637. </div>
  638. </div>
  639. </div>
  640. <div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
  641. <div class="md-sidebar__scrollwrap">
  642. <div class="md-sidebar__inner">
  643. <nav class="md-nav md-nav--secondary" aria-label="Table of contents">
  644. </nav>
  645. </div>
  646. </div>
  647. </div>
  648. <div class="md-content" data-md-component="content">
  649. <article class="md-content__inner md-typeset">
  650. <h1 id="quantization">Quantization</h1>
  651. <p>Quantization is significant to accelerate the model inference. Since there's little accuracy (performance) reduction when quantizing the model, get it easy to quantize it!</p>
  652. <p>To quantize the model, please call <code>Quantize</code> from <code>LLamaQuantizer</code>, which is a static method.</p>
  653. <pre><code class="language-cs">string srcPath = &quot;&lt;model.bin&gt;&quot;;
  654. string dstPath = &quot;&lt;model_q4_0.bin&gt;&quot;;
  655. LLamaQuantizer.Quantize(srcPath, dstPath, &quot;q4_0&quot;);
  656. // The following overload is also okay.
  657. // LLamaQuantizer.Quantize(srcPath, dstPath, LLamaFtype.LLAMA_FTYPE_MOSTLY_Q4_0);
  658. </code></pre>
  659. <p>After calling it, a quantized model file will be saved.</p>
  660. <p>There're currently 5 types of quantization supported:</p>
  661. <ul>
  662. <li>q4_0</li>
  663. <li>q4_1</li>
  664. <li>q5_0</li>
  665. <li>q5_1</li>
  666. <li>q8_0</li>
  667. </ul>
  668. </article>
  669. </div>
  670. </div>
  671. </main>
  672. <footer class="md-footer">
  673. <div class="md-footer-meta md-typeset">
  674. <div class="md-footer-meta__inner md-grid">
  675. <div class="md-copyright">
  676. Made with
  677. <a href="https://squidfunk.github.io/mkdocs-material/" target="_blank" rel="noopener">
  678. Material for MkDocs
  679. </a>
  680. </div>
  681. </div>
  682. </div>
  683. </footer>
  684. </div>
  685. <div class="md-dialog" data-md-component="dialog">
  686. <div class="md-dialog__inner md-typeset"></div>
  687. </div>
  688. <script id="__config" type="application/json">{"base": "../..", "features": [], "search": "../../assets/javascripts/workers/search.208ed371.min.js", "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": {"provider": "mike"}}</script>
  689. <script src="../../assets/javascripts/bundle.a51614de.min.js"></script>
  690. </body>
  691. </html>

C#/.NET上易用的LLM高性能推理框架,支持LLaMA和LLaVA系列模型。

Contributors (1)