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 40 kB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980818283848586878889909192939495969798991001011021031041051061071081091101111121131141151161171181191201211221231241251261271281291301311321331341351361371381391401411421431441451461471481491501511521531541551561571581591601611621631641651661671681691701711721731741751761771781791801811821831841851861871881891901911921931941951961971981992002012022032042052062072082092102112122132142152162172182192202212222232242252262272282292302312322332342352362372382392402412422432442452462472482492502512522532542552562572582592602612622632642652662672682692702712722732742752762772782792802812822832842852862872882892902912922932942952962972982993003013023033043053063073083093103113123133143153163173183193203213223233243253263273283293303313323333343353363373383393403413423433443453463473483493503513523533543553563573583593603613623633643653663673683693703713723733743753763773783793803813823833843853863873883893903913923933943953963973983994004014024034044054064074084094104114124134144154164174184194204214224234244254264274284294304314324334344354364374384394404414424434444454464474484494504514524534544554564574584594604614624634644654664674684694704714724734744754764774784794804814824834844854864874884894904914924934944954964974984995005015025035045055065075085095105115125135145155165175185195205215225235245255265275285295305315325335345355365375385395405415425435445455465475485495505515525535545555565575585595605615625635645655665675685695705715725735745755765775785795805815825835845855865875885895905915925935945955965975985996006016026036046056066076086096106116126136146156166176186196206216226236246256266276286296306316326336346356366376386396406416426436446456466476486496506516526536546556566576586596606616626636646656666676686696706716726736746756766776786796806816826836846856866876886896906916926936946956966976986997007017027037047057067077087097107117127137147157167177187197207217227237247257267277287297307317327337347357367377387397407417427437447457467477487497507517527537547557567577587597607617627637647657667677687697707717727737747757767777787797807817827837847857867877887897907917927937947957967977987998008018028038048058068078088098108118128138148158168178188198208218228238248258268278288298308318328338348358368378388398408418428438448458468478488498508518528538548558568578588598608618628638648658668678688698708718728738748758768778788798808818828838848858868878888898908918928938948958968978988999009019029039049059069079089099109119129139149159169179189199209219229239249259269279289299309319329339349359369379389399409419429439449459469479489499509519529539549559569579589599609619629639649659669679689699709719729739749759769779789799809819829839849859869879889899909919929939949959969979989991000100110021003100410051006100710081009101010111012101310141015101610171018101910201021102210231024102510261027102810291030103110321033103410351036103710381039104010411042104310441045104610471048104910501051105210531054105510561057105810591060106110621063106410651066106710681069107010711072107310741075107610771078107910801081108210831084108510861087108810891090109110921093109410951096109710981099110011011102110311041105110611071108110911101111111211131114111511161117111811191120112111221123112411251126112711281129113011311132113311341135113611371138113911401141114211431144114511461147114811491150115111521153115411551156115711581159116011611162116311641165116611671168116911701171117211731174117511761177117811791180118111821183118411851186118711881189119011911192119311941195119611971198119912001201120212031204120512061207120812091210121112121213121412151216121712181219122012211222122312241225122612271228122912301231123212331234123512361237123812391240124112421243124412451246124712481249125012511252125312541255125612571258125912601261126212631264126512661267126812691270127112721273127412751276127712781279128012811282128312841285128612871288128912901291129212931294129512961297129812991300130113021303130413051306130713081309131013111312131313141315131613171318131913201321132213231324132513261327132813291330133113321333133413351336133713381339134013411342134313441345134613471348134913501351135213531354135513561357135813591360136113621363136413651366136713681369137013711372137313741375137613771378137913801381138213831384138513861387138813891390139113921393139413951396139713981399140014011402140314041405140614071408140914101411141214131414141514161417141814191420142114221423142414251426142714281429143014311432143314341435143614371438143914401441144214431444144514461447144814491450145114521453145414551456145714581459146014611462146314641465146614671468146914701471147214731474147514761477147814791480148114821483148414851486148714881489149014911492149314941495149614971498149915001501150215031504150515061507150815091510151115121513151415151516151715181519152015211522152315241525152615271528152915301531153215331534153515361537153815391540154115421543154415451546154715481549155015511552155315541555155615571558155915601561156215631564156515661567156815691570157115721573157415751576157715781579158015811582158315841585158615871588158915901591159215931594159515961597159815991600160116021603160416051606160716081609161016111612161316141615161616171618161916201621162216231624162516261627162816291630163116321633163416351636163716381639164016411642164316441645164616471648164916501651165216531654165516561657165816591660166116621663166416651666166716681669167016711672167316741675167616771678167916801681168216831684168516861687168816891690169116921693169416951696169716981699170017011702170317041705170617071708170917101711171217131714171517161717171817191720172117221723172417251726172717281729173017311732173317341735173617371738173917401741174217431744174517461747174817491750175117521753175417551756175717581759176017611762176317641765176617671768176917701771177217731774177517761777177817791780178117821783178417851786178717881789179017911792179317941795179617971798179918001801180218031804180518061807180818091810181118121813181418151816181718181819182018211822182318241825182618271828182918301831183218331834183518361837183818391840184118421843184418451846184718481849185018511852185318541855185618571858185918601861186218631864186518661867186818691870187118721873187418751876187718781879188018811882188318841885188618871888188918901891189218931894189518961897189818991900190119021903190419051906190719081909191019111912191319141915191619171918191919201921192219231924192519261927192819291930193119321933193419351936193719381939194019411942194319441945194619471948194919501951195219531954195519561957195819591960196119621963196419651966196719681969197019711972197319741975197619771978197919801981198219831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020202120222023202420252026202720282029203020312032203320342035203620372038203920402041
  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="icon" href="../../assets/images/favicon.png">
  7. <meta name="generator" content="mkdocs-1.4.3, mkdocs-material-9.1.16">
  8. <title>Quantization - LLamaSharp Documentation</title>
  9. <link rel="stylesheet" href="../../assets/stylesheets/main.26e3688c.min.css">
  10. <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  11. <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">
  12. <style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
  13. <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>
  14. </head>
  15. <body dir="ltr">
  16. <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>
  17. <input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
  18. <input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
  19. <label class="md-overlay" for="__drawer"></label>
  20. <div data-md-component="skip">
  21. <a href="#quantization" class="md-skip">
  22. Skip to content
  23. </a>
  24. </div>
  25. <div data-md-component="announce">
  26. </div>
  27. <div data-md-color-scheme="default" data-md-component="outdated" hidden>
  28. </div>
  29. <header class="md-header md-header--shadow" data-md-component="header">
  30. <nav class="md-header__inner md-grid" aria-label="Header">
  31. <a href="../.." title="LLamaSharp Documentation" class="md-header__button md-logo" aria-label="LLamaSharp Documentation" data-md-component="logo">
  32. <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>
  33. </a>
  34. <label class="md-header__button md-icon" for="__drawer">
  35. <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3V6m0 5h18v2H3v-2m0 5h18v2H3v-2Z"/></svg>
  36. </label>
  37. <div class="md-header__title" data-md-component="header-title">
  38. <div class="md-header__ellipsis">
  39. <div class="md-header__topic">
  40. <span class="md-ellipsis">
  41. LLamaSharp Documentation
  42. </span>
  43. </div>
  44. <div class="md-header__topic" data-md-component="header-topic">
  45. <span class="md-ellipsis">
  46. Quantization
  47. </span>
  48. </div>
  49. </div>
  50. </div>
  51. <label class="md-header__button md-icon" for="__search">
  52. <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>
  53. </label>
  54. <div class="md-search" data-md-component="search" role="dialog">
  55. <label class="md-search__overlay" for="__search"></label>
  56. <div class="md-search__inner" role="search">
  57. <form class="md-search__form" name="search">
  58. <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>
  59. <label class="md-search__icon md-icon" for="__search">
  60. <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>
  61. <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>
  62. </label>
  63. <nav class="md-search__options" aria-label="Search">
  64. <button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
  65. <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>
  66. </button>
  67. </nav>
  68. </form>
  69. <div class="md-search__output">
  70. <div class="md-search__scrollwrap" data-md-scrollfix>
  71. <div class="md-search-result" data-md-component="search-result">
  72. <div class="md-search-result__meta">
  73. Initializing search
  74. </div>
  75. <ol class="md-search-result__list" role="presentation"></ol>
  76. </div>
  77. </div>
  78. </div>
  79. </div>
  80. </div>
  81. </nav>
  82. </header>
  83. <div class="md-container" data-md-component="container">
  84. <main class="md-main" data-md-component="main">
  85. <div class="md-main__inner md-grid">
  86. <div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
  87. <div class="md-sidebar__scrollwrap">
  88. <div class="md-sidebar__inner">
  89. <nav class="md-nav md-nav--primary" aria-label="Navigation" data-md-level="0">
  90. <label class="md-nav__title" for="__drawer">
  91. <a href="../.." title="LLamaSharp Documentation" class="md-nav__button md-logo" aria-label="LLamaSharp Documentation" data-md-component="logo">
  92. <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>
  93. </a>
  94. LLamaSharp Documentation
  95. </label>
  96. <ul class="md-nav__list" data-md-scrollfix>
  97. <li class="md-nav__item">
  98. <a href="../.." class="md-nav__link">
  99. Overview
  100. </a>
  101. </li>
  102. <li class="md-nav__item">
  103. <a href="../../GetStarted/" class="md-nav__link">
  104. Get Started
  105. </a>
  106. </li>
  107. <li class="md-nav__item">
  108. <a href="../../Architecture/" class="md-nav__link">
  109. Architecture
  110. </a>
  111. </li>
  112. <li class="md-nav__item">
  113. <a href="../../Tricks/" class="md-nav__link">
  114. Tricks for FAQ
  115. </a>
  116. </li>
  117. <li class="md-nav__item">
  118. <a href="../../ContributingGuide/" class="md-nav__link">
  119. Contributing Guide
  120. </a>
  121. </li>
  122. <li class="md-nav__item md-nav__item--nested">
  123. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_6" >
  124. <label class="md-nav__link" for="__nav_6" id="__nav_6_label" tabindex="0">
  125. LLamaContext
  126. <span class="md-nav__icon md-icon"></span>
  127. </label>
  128. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_6_label" aria-expanded="false">
  129. <label class="md-nav__title" for="__nav_6">
  130. <span class="md-nav__icon md-icon"></span>
  131. LLamaContext
  132. </label>
  133. <ul class="md-nav__list" data-md-scrollfix>
  134. <li class="md-nav__item">
  135. <a href="../../LLamaContext/parameters.md" class="md-nav__link">
  136. Context Parameters
  137. </a>
  138. </li>
  139. <li class="md-nav__item">
  140. <a href="../../LLamaContext/tokenization.md" class="md-nav__link">
  141. Tokenization
  142. </a>
  143. </li>
  144. <li class="md-nav__item">
  145. <a href="../../LLamaContext/embeddings.md" class="md-nav__link">
  146. Get Embeddings
  147. </a>
  148. </li>
  149. <li class="md-nav__item">
  150. <a href="../../LLamaContext/quantization.md" class="md-nav__link">
  151. Quantization
  152. </a>
  153. </li>
  154. <li class="md-nav__item">
  155. <a href="../../LLamaContext/save-load-state.md" class="md-nav__link">
  156. Save/Load State
  157. </a>
  158. </li>
  159. </ul>
  160. </nav>
  161. </li>
  162. <li class="md-nav__item md-nav__item--nested">
  163. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_7" >
  164. <label class="md-nav__link" for="__nav_7" id="__nav_7_label" tabindex="0">
  165. LLamaExecutors
  166. <span class="md-nav__icon md-icon"></span>
  167. </label>
  168. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_7_label" aria-expanded="false">
  169. <label class="md-nav__title" for="__nav_7">
  170. <span class="md-nav__icon md-icon"></span>
  171. LLamaExecutors
  172. </label>
  173. <ul class="md-nav__list" data-md-scrollfix>
  174. <li class="md-nav__item">
  175. <a href="../../LLamaExecutors/parameters/" class="md-nav__link">
  176. Inference Parameters
  177. </a>
  178. </li>
  179. <li class="md-nav__item">
  180. <a href="../../LLamaExecutors/text-to-text-apis/" class="md-nav__link">
  181. Text-to-Text APIs
  182. </a>
  183. </li>
  184. <li class="md-nav__item">
  185. <a href="../../LLamaExecutors/save-load-state/" class="md-nav__link">
  186. Save/Load State
  187. </a>
  188. </li>
  189. <li class="md-nav__item">
  190. <a href="../../LLamaExecutors/differences/" class="md-nav__link">
  191. Differences of Executors
  192. </a>
  193. </li>
  194. </ul>
  195. </nav>
  196. </li>
  197. <li class="md-nav__item md-nav__item--nested">
  198. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_8" >
  199. <label class="md-nav__link" for="__nav_8" id="__nav_8_label" tabindex="0">
  200. ChatSession
  201. <span class="md-nav__icon md-icon"></span>
  202. </label>
  203. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_8_label" aria-expanded="false">
  204. <label class="md-nav__title" for="__nav_8">
  205. <span class="md-nav__icon md-icon"></span>
  206. ChatSession
  207. </label>
  208. <ul class="md-nav__list" data-md-scrollfix>
  209. <li class="md-nav__item">
  210. <a href="../../ChatSession/basic-usages/" class="md-nav__link">
  211. Basic Usages
  212. </a>
  213. </li>
  214. <li class="md-nav__item">
  215. <a href="../../ChatSession/transforms/" class="md-nav__link">
  216. Transoforms
  217. </a>
  218. </li>
  219. <li class="md-nav__item">
  220. <a href="../../ChatSession/save-load-session/" class="md-nav__link">
  221. Save/Load Session
  222. </a>
  223. </li>
  224. </ul>
  225. </nav>
  226. </li>
  227. <li class="md-nav__item md-nav__item--nested">
  228. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_9" >
  229. <label class="md-nav__link" for="__nav_9" id="__nav_9_label" tabindex="0">
  230. Non-English Usages
  231. <span class="md-nav__icon md-icon"></span>
  232. </label>
  233. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_9_label" aria-expanded="false">
  234. <label class="md-nav__title" for="__nav_9">
  235. <span class="md-nav__icon md-icon"></span>
  236. Non-English Usages
  237. </label>
  238. <ul class="md-nav__list" data-md-scrollfix>
  239. <li class="md-nav__item">
  240. <a href="../../NonEnglishUsage/Chinese/" class="md-nav__link">
  241. Chinese
  242. </a>
  243. </li>
  244. </ul>
  245. </nav>
  246. </li>
  247. <li class="md-nav__item md-nav__item--nested">
  248. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_10" >
  249. <label class="md-nav__link" for="__nav_10" id="__nav_10_label" tabindex="0">
  250. High-level Applications
  251. <span class="md-nav__icon md-icon"></span>
  252. </label>
  253. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_10_label" aria-expanded="false">
  254. <label class="md-nav__title" for="__nav_10">
  255. <span class="md-nav__icon md-icon"></span>
  256. High-level Applications
  257. </label>
  258. <ul class="md-nav__list" data-md-scrollfix>
  259. <li class="md-nav__item">
  260. <a href="../../HighLevelApps/bot-sharp/" class="md-nav__link">
  261. BotSharp
  262. </a>
  263. </li>
  264. <li class="md-nav__item">
  265. <a href="../../HighLevelApps/semantic-kernel/" class="md-nav__link">
  266. semantic-kernel
  267. </a>
  268. </li>
  269. </ul>
  270. </nav>
  271. </li>
  272. <li class="md-nav__item md-nav__item--nested">
  273. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_11" >
  274. <label class="md-nav__link" for="__nav_11" id="__nav_11_label" tabindex="0">
  275. More
  276. <span class="md-nav__icon md-icon"></span>
  277. </label>
  278. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_11_label" aria-expanded="false">
  279. <label class="md-nav__title" for="__nav_11">
  280. <span class="md-nav__icon md-icon"></span>
  281. More
  282. </label>
  283. <ul class="md-nav__list" data-md-scrollfix>
  284. <li class="md-nav__item">
  285. <a href="../../More/log/" class="md-nav__link">
  286. Logger
  287. </a>
  288. </li>
  289. </ul>
  290. </nav>
  291. </li>
  292. <li class="md-nav__item md-nav__item--nested">
  293. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_12" >
  294. <label class="md-nav__link" for="__nav_12" id="__nav_12_label" tabindex="0">
  295. Examples
  296. <span class="md-nav__icon md-icon"></span>
  297. </label>
  298. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_12_label" aria-expanded="false">
  299. <label class="md-nav__title" for="__nav_12">
  300. <span class="md-nav__icon md-icon"></span>
  301. Examples
  302. </label>
  303. <ul class="md-nav__list" data-md-scrollfix>
  304. <li class="md-nav__item">
  305. <a href="../../Examples/ChatSessionStripRoleName/" class="md-nav__link">
  306. Chat session 1
  307. </a>
  308. </li>
  309. <li class="md-nav__item">
  310. <a href="../../Examples/ChatSessionWithRoleName/" class="md-nav__link">
  311. Chat session 2
  312. </a>
  313. </li>
  314. <li class="md-nav__item">
  315. <a href="../../Examples/GetEmbeddings/" class="md-nav__link">
  316. Get embeddings
  317. </a>
  318. </li>
  319. <li class="md-nav__item">
  320. <a href="../../Examples/InstructModeExecute/" class="md-nav__link">
  321. Instruct executor
  322. </a>
  323. </li>
  324. <li class="md-nav__item">
  325. <a href="../../Examples/InteractiveModeExecute/" class="md-nav__link">
  326. Interactive executor
  327. </a>
  328. </li>
  329. <li class="md-nav__item">
  330. <a href="../../Examples/StatelessModeExecute/" class="md-nav__link">
  331. Stateless exeutor
  332. </a>
  333. </li>
  334. <li class="md-nav__item">
  335. <a href="../../Examples/LoadAndSaveSession/" class="md-nav__link">
  336. Load/Save session
  337. </a>
  338. </li>
  339. <li class="md-nav__item">
  340. <a href="../../Examples/LoadAndSaveState/" class="md-nav__link">
  341. Load/Save state
  342. </a>
  343. </li>
  344. <li class="md-nav__item">
  345. <a href="../../Examples/QuantizeModel/" class="md-nav__link">
  346. Quantize model
  347. </a>
  348. </li>
  349. </ul>
  350. </nav>
  351. </li>
  352. <li class="md-nav__item md-nav__item--nested">
  353. <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_13" >
  354. <label class="md-nav__link" for="__nav_13" id="__nav_13_label" tabindex="0">
  355. API Reference
  356. <span class="md-nav__icon md-icon"></span>
  357. </label>
  358. <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_13_label" aria-expanded="false">
  359. <label class="md-nav__title" for="__nav_13">
  360. <span class="md-nav__icon md-icon"></span>
  361. API Reference
  362. </label>
  363. <ul class="md-nav__list" data-md-scrollfix>
  364. <li class="md-nav__item">
  365. <a href="../../xmldocs/" class="md-nav__link">
  366. index
  367. </a>
  368. </li>
  369. <li class="md-nav__item">
  370. <a href="../../xmldocs/llama.abstractions.ihistorytransform/" class="md-nav__link">
  371. llama.abstractions.ihistorytransform
  372. </a>
  373. </li>
  374. <li class="md-nav__item">
  375. <a href="../../xmldocs/llama.abstractions.iinferenceparams/" class="md-nav__link">
  376. llama.abstractions.iinferenceparams
  377. </a>
  378. </li>
  379. <li class="md-nav__item">
  380. <a href="../../xmldocs/llama.abstractions.illamaexecutor/" class="md-nav__link">
  381. llama.abstractions.illamaexecutor
  382. </a>
  383. </li>
  384. <li class="md-nav__item">
  385. <a href="../../xmldocs/llama.abstractions.imodelparams/" class="md-nav__link">
  386. llama.abstractions.imodelparams
  387. </a>
  388. </li>
  389. <li class="md-nav__item">
  390. <a href="../../xmldocs/llama.abstractions.itextstreamtransform/" class="md-nav__link">
  391. llama.abstractions.itextstreamtransform
  392. </a>
  393. </li>
  394. <li class="md-nav__item">
  395. <a href="../../xmldocs/llama.abstractions.itexttransform/" class="md-nav__link">
  396. llama.abstractions.itexttransform
  397. </a>
  398. </li>
  399. <li class="md-nav__item">
  400. <a href="../../xmldocs/llama.chatsession/" class="md-nav__link">
  401. llama.chatsession
  402. </a>
  403. </li>
  404. <li class="md-nav__item">
  405. <a href="../../xmldocs/llama.common.authorrole/" class="md-nav__link">
  406. llama.common.authorrole
  407. </a>
  408. </li>
  409. <li class="md-nav__item">
  410. <a href="../../xmldocs/llama.common.chathistory/" class="md-nav__link">
  411. llama.common.chathistory
  412. </a>
  413. </li>
  414. <li class="md-nav__item">
  415. <a href="../../xmldocs/llama.common.fixedsizequeue-1/" class="md-nav__link">
  416. llama.common.fixedsizequeue-1
  417. </a>
  418. </li>
  419. <li class="md-nav__item">
  420. <a href="../../xmldocs/llama.common.illamalogger/" class="md-nav__link">
  421. llama.common.illamalogger
  422. </a>
  423. </li>
  424. <li class="md-nav__item">
  425. <a href="../../xmldocs/llama.common.inferenceparams/" class="md-nav__link">
  426. llama.common.inferenceparams
  427. </a>
  428. </li>
  429. <li class="md-nav__item">
  430. <a href="../../xmldocs/llama.common.llamadefaultlogger/" class="md-nav__link">
  431. llama.common.llamadefaultlogger
  432. </a>
  433. </li>
  434. <li class="md-nav__item">
  435. <a href="../../xmldocs/llama.common.mirostattype/" class="md-nav__link">
  436. llama.common.mirostattype
  437. </a>
  438. </li>
  439. <li class="md-nav__item">
  440. <a href="../../xmldocs/llama.common.modelparams/" class="md-nav__link">
  441. llama.common.modelparams
  442. </a>
  443. </li>
  444. <li class="md-nav__item">
  445. <a href="../../xmldocs/llama.exceptions.grammarexpectedname/" class="md-nav__link">
  446. llama.exceptions.grammarexpectedname
  447. </a>
  448. </li>
  449. <li class="md-nav__item">
  450. <a href="../../xmldocs/llama.exceptions.grammarexpectednext/" class="md-nav__link">
  451. llama.exceptions.grammarexpectednext
  452. </a>
  453. </li>
  454. <li class="md-nav__item">
  455. <a href="../../xmldocs/llama.exceptions.grammarexpectedprevious/" class="md-nav__link">
  456. llama.exceptions.grammarexpectedprevious
  457. </a>
  458. </li>
  459. <li class="md-nav__item">
  460. <a href="../../xmldocs/llama.exceptions.grammarformatexception/" class="md-nav__link">
  461. llama.exceptions.grammarformatexception
  462. </a>
  463. </li>
  464. <li class="md-nav__item">
  465. <a href="../../xmldocs/llama.exceptions.grammarunexpectedcharaltelement/" class="md-nav__link">
  466. llama.exceptions.grammarunexpectedcharaltelement
  467. </a>
  468. </li>
  469. <li class="md-nav__item">
  470. <a href="../../xmldocs/llama.exceptions.grammarunexpectedcharrngelement/" class="md-nav__link">
  471. llama.exceptions.grammarunexpectedcharrngelement
  472. </a>
  473. </li>
  474. <li class="md-nav__item">
  475. <a href="../../xmldocs/llama.exceptions.grammarunexpectedendelement/" class="md-nav__link">
  476. llama.exceptions.grammarunexpectedendelement
  477. </a>
  478. </li>
  479. <li class="md-nav__item">
  480. <a href="../../xmldocs/llama.exceptions.grammarunexpectedendofinput/" class="md-nav__link">
  481. llama.exceptions.grammarunexpectedendofinput
  482. </a>
  483. </li>
  484. <li class="md-nav__item">
  485. <a href="../../xmldocs/llama.exceptions.grammarunexpectedhexcharscount/" class="md-nav__link">
  486. llama.exceptions.grammarunexpectedhexcharscount
  487. </a>
  488. </li>
  489. <li class="md-nav__item">
  490. <a href="../../xmldocs/llama.exceptions.grammarunknownescapecharacter/" class="md-nav__link">
  491. llama.exceptions.grammarunknownescapecharacter
  492. </a>
  493. </li>
  494. <li class="md-nav__item">
  495. <a href="../../xmldocs/llama.exceptions.runtimeerror/" class="md-nav__link">
  496. llama.exceptions.runtimeerror
  497. </a>
  498. </li>
  499. <li class="md-nav__item">
  500. <a href="../../xmldocs/llama.extensions.imodelparamsextensions/" class="md-nav__link">
  501. llama.extensions.imodelparamsextensions
  502. </a>
  503. </li>
  504. <li class="md-nav__item">
  505. <a href="../../xmldocs/llama.extensions.keyvaluepairextensions/" class="md-nav__link">
  506. llama.extensions.keyvaluepairextensions
  507. </a>
  508. </li>
  509. <li class="md-nav__item">
  510. <a href="../../xmldocs/llama.grammars.grammar/" class="md-nav__link">
  511. llama.grammars.grammar
  512. </a>
  513. </li>
  514. <li class="md-nav__item">
  515. <a href="../../xmldocs/llama.grammars.grammarrule/" class="md-nav__link">
  516. llama.grammars.grammarrule
  517. </a>
  518. </li>
  519. <li class="md-nav__item">
  520. <a href="../../xmldocs/llama.instructexecutor/" class="md-nav__link">
  521. llama.instructexecutor
  522. </a>
  523. </li>
  524. <li class="md-nav__item">
  525. <a href="../../xmldocs/llama.interactiveexecutor/" class="md-nav__link">
  526. llama.interactiveexecutor
  527. </a>
  528. </li>
  529. <li class="md-nav__item">
  530. <a href="../../xmldocs/llama.llamacontext/" class="md-nav__link">
  531. llama.llamacontext
  532. </a>
  533. </li>
  534. <li class="md-nav__item">
  535. <a href="../../xmldocs/llama.llamaembedder/" class="md-nav__link">
  536. llama.llamaembedder
  537. </a>
  538. </li>
  539. <li class="md-nav__item">
  540. <a href="../../xmldocs/llama.llamaquantizer/" class="md-nav__link">
  541. llama.llamaquantizer
  542. </a>
  543. </li>
  544. <li class="md-nav__item">
  545. <a href="../../xmldocs/llama.llamatransforms/" class="md-nav__link">
  546. llama.llamatransforms
  547. </a>
  548. </li>
  549. <li class="md-nav__item">
  550. <a href="../../xmldocs/llama.llamaweights/" class="md-nav__link">
  551. llama.llamaweights
  552. </a>
  553. </li>
  554. <li class="md-nav__item">
  555. <a href="../../xmldocs/llama.native.llamacontextparams/" class="md-nav__link">
  556. llama.native.llamacontextparams
  557. </a>
  558. </li>
  559. <li class="md-nav__item">
  560. <a href="../../xmldocs/llama.native.llamaftype/" class="md-nav__link">
  561. llama.native.llamaftype
  562. </a>
  563. </li>
  564. <li class="md-nav__item">
  565. <a href="../../xmldocs/llama.native.llamagrammarelement/" class="md-nav__link">
  566. llama.native.llamagrammarelement
  567. </a>
  568. </li>
  569. <li class="md-nav__item">
  570. <a href="../../xmldocs/llama.native.llamagrammarelementtype/" class="md-nav__link">
  571. llama.native.llamagrammarelementtype
  572. </a>
  573. </li>
  574. <li class="md-nav__item">
  575. <a href="../../xmldocs/llama.native.llamamodelquantizeparams/" class="md-nav__link">
  576. llama.native.llamamodelquantizeparams
  577. </a>
  578. </li>
  579. <li class="md-nav__item">
  580. <a href="../../xmldocs/llama.native.llamatokendata/" class="md-nav__link">
  581. llama.native.llamatokendata
  582. </a>
  583. </li>
  584. <li class="md-nav__item">
  585. <a href="../../xmldocs/llama.native.llamatokendataarray/" class="md-nav__link">
  586. llama.native.llamatokendataarray
  587. </a>
  588. </li>
  589. <li class="md-nav__item">
  590. <a href="../../xmldocs/llama.native.llamatokendataarraynative/" class="md-nav__link">
  591. llama.native.llamatokendataarraynative
  592. </a>
  593. </li>
  594. <li class="md-nav__item">
  595. <a href="../../xmldocs/llama.native.nativeapi/" class="md-nav__link">
  596. llama.native.nativeapi
  597. </a>
  598. </li>
  599. <li class="md-nav__item">
  600. <a href="../../xmldocs/llama.native.safellamacontexthandle/" class="md-nav__link">
  601. llama.native.safellamacontexthandle
  602. </a>
  603. </li>
  604. <li class="md-nav__item">
  605. <a href="../../xmldocs/llama.native.safellamagrammarhandle/" class="md-nav__link">
  606. llama.native.safellamagrammarhandle
  607. </a>
  608. </li>
  609. <li class="md-nav__item">
  610. <a href="../../xmldocs/llama.native.safellamahandlebase/" class="md-nav__link">
  611. llama.native.safellamahandlebase
  612. </a>
  613. </li>
  614. <li class="md-nav__item">
  615. <a href="../../xmldocs/llama.native.safellamamodelhandle/" class="md-nav__link">
  616. llama.native.safellamamodelhandle
  617. </a>
  618. </li>
  619. <li class="md-nav__item">
  620. <a href="../../xmldocs/llama.native.samplingapi/" class="md-nav__link">
  621. llama.native.samplingapi
  622. </a>
  623. </li>
  624. <li class="md-nav__item">
  625. <a href="../../xmldocs/llama.oldversion.chatcompletion/" class="md-nav__link">
  626. llama.oldversion.chatcompletion
  627. </a>
  628. </li>
  629. <li class="md-nav__item">
  630. <a href="../../xmldocs/llama.oldversion.chatcompletionchoice/" class="md-nav__link">
  631. llama.oldversion.chatcompletionchoice
  632. </a>
  633. </li>
  634. <li class="md-nav__item">
  635. <a href="../../xmldocs/llama.oldversion.chatcompletionchunk/" class="md-nav__link">
  636. llama.oldversion.chatcompletionchunk
  637. </a>
  638. </li>
  639. <li class="md-nav__item">
  640. <a href="../../xmldocs/llama.oldversion.chatcompletionchunkchoice/" class="md-nav__link">
  641. llama.oldversion.chatcompletionchunkchoice
  642. </a>
  643. </li>
  644. <li class="md-nav__item">
  645. <a href="../../xmldocs/llama.oldversion.chatcompletionchunkdelta/" class="md-nav__link">
  646. llama.oldversion.chatcompletionchunkdelta
  647. </a>
  648. </li>
  649. <li class="md-nav__item">
  650. <a href="../../xmldocs/llama.oldversion.chatcompletionmessage/" class="md-nav__link">
  651. llama.oldversion.chatcompletionmessage
  652. </a>
  653. </li>
  654. <li class="md-nav__item">
  655. <a href="../../xmldocs/llama.oldversion.chatmessagerecord/" class="md-nav__link">
  656. llama.oldversion.chatmessagerecord
  657. </a>
  658. </li>
  659. <li class="md-nav__item">
  660. <a href="../../xmldocs/llama.oldversion.chatrole/" class="md-nav__link">
  661. llama.oldversion.chatrole
  662. </a>
  663. </li>
  664. <li class="md-nav__item">
  665. <a href="../../xmldocs/llama.oldversion.chatsession-1/" class="md-nav__link">
  666. llama.oldversion.chatsession-1
  667. </a>
  668. </li>
  669. <li class="md-nav__item">
  670. <a href="../../xmldocs/llama.oldversion.completion/" class="md-nav__link">
  671. llama.oldversion.completion
  672. </a>
  673. </li>
  674. <li class="md-nav__item">
  675. <a href="../../xmldocs/llama.oldversion.completionchoice/" class="md-nav__link">
  676. llama.oldversion.completionchoice
  677. </a>
  678. </li>
  679. <li class="md-nav__item">
  680. <a href="../../xmldocs/llama.oldversion.completionchunk/" class="md-nav__link">
  681. llama.oldversion.completionchunk
  682. </a>
  683. </li>
  684. <li class="md-nav__item">
  685. <a href="../../xmldocs/llama.oldversion.completionlogprobs/" class="md-nav__link">
  686. llama.oldversion.completionlogprobs
  687. </a>
  688. </li>
  689. <li class="md-nav__item">
  690. <a href="../../xmldocs/llama.oldversion.completionusage/" class="md-nav__link">
  691. llama.oldversion.completionusage
  692. </a>
  693. </li>
  694. <li class="md-nav__item">
  695. <a href="../../xmldocs/llama.oldversion.embedding/" class="md-nav__link">
  696. llama.oldversion.embedding
  697. </a>
  698. </li>
  699. <li class="md-nav__item">
  700. <a href="../../xmldocs/llama.oldversion.embeddingdata/" class="md-nav__link">
  701. llama.oldversion.embeddingdata
  702. </a>
  703. </li>
  704. <li class="md-nav__item">
  705. <a href="../../xmldocs/llama.oldversion.embeddingusage/" class="md-nav__link">
  706. llama.oldversion.embeddingusage
  707. </a>
  708. </li>
  709. <li class="md-nav__item">
  710. <a href="../../xmldocs/llama.oldversion.ichatmodel/" class="md-nav__link">
  711. llama.oldversion.ichatmodel
  712. </a>
  713. </li>
  714. <li class="md-nav__item">
  715. <a href="../../xmldocs/llama.oldversion.llamaembedder/" class="md-nav__link">
  716. llama.oldversion.llamaembedder
  717. </a>
  718. </li>
  719. <li class="md-nav__item">
  720. <a href="../../xmldocs/llama.oldversion.llamamodel/" class="md-nav__link">
  721. llama.oldversion.llamamodel
  722. </a>
  723. </li>
  724. <li class="md-nav__item">
  725. <a href="../../xmldocs/llama.oldversion.llamaparams/" class="md-nav__link">
  726. llama.oldversion.llamaparams
  727. </a>
  728. </li>
  729. <li class="md-nav__item">
  730. <a href="../../xmldocs/llama.statefulexecutorbase/" class="md-nav__link">
  731. llama.statefulexecutorbase
  732. </a>
  733. </li>
  734. <li class="md-nav__item">
  735. <a href="../../xmldocs/llama.statelessexecutor/" class="md-nav__link">
  736. llama.statelessexecutor
  737. </a>
  738. </li>
  739. <li class="md-nav__item">
  740. <a href="../../xmldocs/llama.utils/" class="md-nav__link">
  741. llama.utils
  742. </a>
  743. </li>
  744. </ul>
  745. </nav>
  746. </li>
  747. </ul>
  748. </nav>
  749. </div>
  750. </div>
  751. </div>
  752. <div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
  753. <div class="md-sidebar__scrollwrap">
  754. <div class="md-sidebar__inner">
  755. <nav class="md-nav md-nav--secondary" aria-label="Table of contents">
  756. </nav>
  757. </div>
  758. </div>
  759. </div>
  760. <div class="md-content" data-md-component="content">
  761. <article class="md-content__inner md-typeset">
  762. <h1 id="quantization">Quantization</h1>
  763. <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>
  764. <p>To quantize the model, please call <code>Quantize</code> from <code>LLamaQuantizer</code>, which is a static method.</p>
  765. <pre><code class="language-cs">string srcPath = &quot;&lt;model.bin&gt;&quot;;
  766. string dstPath = &quot;&lt;model_q4_0.bin&gt;&quot;;
  767. LLamaQuantizer.Quantize(srcPath, dstPath, &quot;q4_0&quot;);
  768. // The following overload is also okay.
  769. // LLamaQuantizer.Quantize(srcPath, dstPath, LLamaFtype.LLAMA_FTYPE_MOSTLY_Q4_0);
  770. </code></pre>
  771. <p>After calling it, a quantized model file will be saved.</p>
  772. <p>There're currently 5 types of quantization supported:</p>
  773. <ul>
  774. <li>q4_0</li>
  775. <li>q4_1</li>
  776. <li>q5_0</li>
  777. <li>q5_1</li>
  778. <li>q8_0</li>
  779. </ul>
  780. </article>
  781. </div>
  782. </div>
  783. </main>
  784. <footer class="md-footer">
  785. <div class="md-footer-meta md-typeset">
  786. <div class="md-footer-meta__inner md-grid">
  787. <div class="md-copyright">
  788. Made with
  789. <a href="https://squidfunk.github.io/mkdocs-material/" target="_blank" rel="noopener">
  790. Material for MkDocs
  791. </a>
  792. </div>
  793. </div>
  794. </div>
  795. </footer>
  796. </div>
  797. <div class="md-dialog" data-md-component="dialog">
  798. <div class="md-dialog__inner md-typeset"></div>
  799. </div>
  800. <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>
  801. <script src="../../assets/javascripts/bundle.a51614de.min.js"></script>
  802. </body>
  803. </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)