optimum-neuron/v0.4.4/en/_app/immutable/nodes/40.84e0ecc0.js

import{s as vt,o as bt,n as lt}from"../chunks/scheduler.56725da7.js";import{S as _t,i as Mt,e as l,s as o,c as u,h as Ct,a as d,d as i,b as s,f as kt,g as p,j as f,k as Lt,l as Et,m as a,n as g,t as x,o as h,p as $}from"../chunks/index.18a26576.js";import{T as st}from"../chunks/Tip.5b941656.js";import{C as Pt}from"../chunks/CopyLLMTxtMenu.4513c8ed.js";import{H as z}from"../chunks/MermaidChart.svelte_svelte_type_style_lang.049405bf.js";function Ht(T){let n,c='If you need to add support for a custom model not listed above, check out our <a href="./contribute/contribute_for_training">contribute for training guide</a> to learn how to implement custom modeling with distributed training support. You can also open an issue in the <a href="https://github.com/huggingface/optimum-neuron/issues" rel="nofollow">Optimum Neuron GitHub repository</a> to request support for it.';return{c(){n=l("p"),n.innerHTML=c},l(r){n=d(r,"P",{"data-svelte-h":!0}),f(n)!=="svelte-1ottfjl"&&(n.innerHTML=c)},m(r,m){a(r,n,m)},p:lt,d(r){r&&i(n)}}}function St(T){let n,c='If a LLM is listed, e.g. a model with a <code>text-generation</code> task, it means that there is also <a href="https://huggingface.co/docs/text-generation-inference/en/index" rel="nofollow">TGI</a> support for it.';return{c(){n=l("p"),n.innerHTML=c},l(r){n=d(r,"P",{"data-svelte-h":!0}),f(n)!=="svelte-26e481"&&(n.innerHTML=c)},m(r,m){a(r,n,m)},p:lt,d(r){r&&i(n)}}}function yt(T){let n,c='To learn how to export a model for inference, you can check this <a href="https://huggingface.co/docs/optimum-neuron/guides/export_model#selecting-a-task" rel="nofollow">guide</a>.';return{c(){n=l("p"),n.innerHTML=c},l(r){n=d(r,"P",{"data-svelte-h":!0}),f(n)!=="svelte-kd3kkz"&&(n.innerHTML=c)},m(r,m){a(r,n,m)},p:lt,d(r){r&&i(n)}}}function Bt(T){let n,c,r,m,v,F,b,G,_,V,M,dt="Training on AWS Trainium instances (Trn1) enables large-scale model training with distributed parallelism strategies.",X,C,ct="<strong>Requirements:</strong>",Q,E,ft="<li>Model must be compatible with the Neuron SDK. If it small enough to fit within 16GB, training is supported for any architecture that can be successfully compiled.</li> <li><strong>Memory constraint:</strong> Each accelerator has 16GB of memory for model weights, gradients, optimizer states, and activations.</li> <li><strong>For large models:</strong> Custom modeling implementation with tensor parallelism and/or pipeline parallelism support is required.</li>",U,P,mt="The following architectures have custom modeling implementations with distributed training support:",W,H,ut="<thead><tr><th>Architecture</th> <th>Task</th> <th>Tensor Parallelism</th> <th>Pipeline Parallelism</th></tr></thead> <tbody><tr><td>Llama, Llama 2, Llama 3</td> <td>text-generation</td> <td>✓</td> <td>✓</td></tr> <tr><td>Qwen3</td> <td>text-generation</td> <td>✓</td> <td>✓</td></tr> <tr><td>Granite</td> <td>text-generation</td> <td>✓</td> <td>✗</td></tr></tbody>",j,w,K,S,O,y,pt="The following table lists the architectures and tasks that Optimum Neuron supports for inference on Amazon EC2 Inf2 instances.",Y,k,J,B,Z,q,gt="<thead><tr><th>Architecture</th> <th>Task</th></tr></thead> <tbody><tr><td>ALBERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>AST</td> <td>feature-extraction, audio-classification</td></tr> <tr><td>BERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>Beit</td> <td>feature-extraction, image-classification</td></tr> <tr><td>CamemBERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>CLIP</td> <td>feature-extraction, image-classification</td></tr> <tr><td>ConvBERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>ConvNext</td> <td>feature-extraction, image-classification</td></tr> <tr><td>ConvNextV2</td> <td>feature-extraction, image-classification</td></tr> <tr><td>CvT</td> <td>feature-extraction, image-classification</td></tr> <tr><td>DeBERTa (INF2 only)</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>DeBERTa-v2 (INF2 only)</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>Deit</td> <td>feature-extraction, image-classification</td></tr> <tr><td>DistilBERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>DonutSwin</td> <td>feature-extraction</td></tr> <tr><td>Dpt</td> <td>feature-extraction</td></tr> <tr><td>ELECTRA</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>ESM</td> <td>feature-extraction, fill-mask, text-classification, token-classification</td></tr> <tr><td>FlauBERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>Granite</td> <td>text-generation</td></tr> <tr><td>Hubert</td> <td>feature-extraction, automatic-speech-recognition, audio-classification</td></tr> <tr><td>Levit</td> <td>feature-extraction, image-classification</td></tr> <tr><td>Llama, Llama 2, Llama 3</td> <td>text-generation</td></tr> <tr><td>Llama 4</td> <td>text-generation</td></tr> <tr><td>Mixtral</td> <td>text-generation</td></tr> <tr><td>MobileBERT</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>MobileNetV2</td> <td>feature-extraction, image-classification, semantic-segmentation</td></tr> <tr><td>MobileViT</td> <td>feature-extraction, image-classification, semantic-segmentation</td></tr> <tr><td>ModernBERT</td> <td>feature-extraction, fill-mask, text-classification, token-classification</td></tr> <tr><td>MPNet</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>Phi3</td> <td>text-generation</td></tr> <tr><td>Phi</td> <td>feature-extraction, text-classification, token-classification</td></tr> <tr><td>Qwen2</td> <td>text-generation</td></tr> <tr><td>Qwen3</td> <td>feature-extraction, text-generation</td></tr> <tr><td>Qwen3Moe</td> <td>text-generation</td></tr> <tr><td>RoBERTa</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>RoFormer</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>SmolLM3</td> <td>text-generation</td></tr> <tr><td>Swin</td> <td>feature-extraction, image-classification</td></tr> <tr><td>T5</td> <td>text2text-generation</td></tr> <tr><td>UniSpeech</td> <td>feature-extraction, automatic-speech-recognition, audio-classification</td></tr> <tr><td>UniSpeech-SAT</td> <td>feature-extraction, automatic-speech-recognition, audio-classification, audio-frame-classification, audio-xvector</td></tr> <tr><td>ViT</td> <td>feature-extraction, image-classification</td></tr> <tr><td>Wav2Vec2</td> <td>feature-extraction, automatic-speech-recognition, audio-classification, audio-frame-classification, audio-xvector</td></tr> <tr><td>WavLM</td> <td>feature-extraction, automatic-speech-recognition, audio-classification, audio-frame-classification, audio-xvector</td></tr> <tr><td>Whisper</td> <td>automatic-speech-recognition</td></tr> <tr><td>XLM</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>XLM-RoBERTa</td> <td>feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification</td></tr> <tr><td>Yolos</td> <td>feature-extraction, object-detection</td></tr></tbody>",tt,R,et,A,xt="<thead><tr><th>Architecture</th> <th>Task</th></tr></thead> <tbody><tr><td>Stable Diffusion</td> <td>text-to-image, image-to-image, inpaint</td></tr> <tr><td>Stable Diffusion XL Base</td> <td>text-to-image, image-to-image, inpaint</td></tr> <tr><td>Stable Diffusion XL Refiner</td> <td>image-to-image, inpaint</td></tr> <tr><td>SDXL Turbo</td> <td>text-to-image, image-to-image, inpaint</td></tr> <tr><td>LCM</td> <td>text-to-image</td></tr> <tr><td>PixArt-α</td> <td>text-to-image</td></tr> <tr><td>PixArt-Σ</td> <td>text-to-image</td></tr> <tr><td>Flux</td> <td>text-to-image, inpaint</td></tr> <tr><td>Flux Kontext</td> <td>text-to-image, image-to-image</td></tr></tbody>",it,D,at,I,ht="<thead><tr><th>Architecture</th> <th>Task</th></tr></thead> <tbody><tr><td>Transformer</td> <td>feature-extraction, sentence-similarity</td></tr> <tr><td>CLIP</td> <td>feature-extraction, zero-shot-image-classification</td></tr></tbody>",nt,L,rt,N,ot;return v=new Pt({props:{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"}}),b=new z({props:{title:"Supported architectures",local:"supported-architectures",headingTag:"h1"}}),_=new z({props:{title:"Training",local:"training",headingTag:"h2"}}),w=new st({props:{$$slots:{default:[Ht]},$$scope:{ctx:T}}}),S=new z({props:{title:"Inference",local:"inference",headingTag:"h2"}}),k=new st({props:{$$slots:{default:[St]},$$scope:{ctx:T}}}),B=new z({props:{title:"Transformers",local:"transformers",headingTag:"h3"}}),R=new z({props:{title:"Diffusers",local:"diffusers",headingTag:"h3"}}),D=new z({props:{title:"Sentence Transformers",local:"sentence-transformers",headingTag:"h3"}}),L=new st({props:{$$slots:{default:[yt]},$$scope:{ctx:T}}}),{c(){n=l("meta"),c=o(),r=l("p"),m=o(),u(v.$$.fragment),F=o(),u(b.$$.fragment),G=o(),u(_.$$.fragment),V=o(),M=l("p"),M.textContent=dt,X=o(),C=l("p"),C.innerHTML=ct,Q=o(),E=l("ul"),E.innerHTML=ft,U=o(),P=l("p"),P.textContent=mt,W=o(),H=l("table"),H.innerHTML=ut,j=o(),u(w.$$.fragment),K=o(),u(S.$$.fragment),O=o(),y=l("p"),y.textContent=pt,Y=o(),u(k.$$.fragment),J=o(),u(B.$$.fragment),Z=o(),q=l("table"),q.innerHTML=gt,tt=o(),u(R.$$.fragment),et=o(),A=l("table"),A.innerHTML=xt,it=o(),u(D.$$.fragment),at=o(),I=l("table"),I.innerHTML=ht,nt=o(),u(L.$$.fragment),rt=o(),N=l("p"),this.h()},l(t){const e=Ct("svelte-u9bgzb",document.head);n=d(e,"META",{name:!0,content:!0}),e.forEach(i),c=s(t),r=d(t,"P",{}),kt(r).forEach(i),m=s(t),p(v.$$.fragment,t),F=s(t),p(b.$$.fragment,t),G=s(t),p(_.$$.fragment,t),V=s(t),M=d(t,"P",{"data-svelte-h":!0}),f(M)!=="svelte-bs045o"&&(M.textContent=dt),X=s(t),C=d(t,"P",{"data-svelte-h":!0}),f(C)!=="svelte-11lzvnb"&&(C.innerHTML=ct),Q=s(t),E=d(t,"UL",{"data-svelte-h":!0}),f(E)!=="svelte-86fnwf"&&(E.innerHTML=ft),U=s(t),P=d(t,"P",{"data-svelte-h":!0}),f(P)!=="svelte-1x9hhfd"&&(P.textContent=mt),W=s(t),H=d(t,"TABLE",{"data-svelte-h":!0}),f(H)!=="svelte-1r98o4n"&&(H.innerHTML=ut),j=s(t),p(w.$$.fragment,t),K=s(t),p(S.$$.fragment,t),O=s(t),y=d(t,"P",{"data-svelte-h":!0}),f(y)!=="svelte-11aox24"&&(y.textContent=pt),Y=s(t),p(k.$$.fragment,t),J=s(t),p(B.$$.fragment,t),Z=s(t),q=d(t,"TABLE",{"data-svelte-h":!0}),f(q)!=="svelte-14vt7nu"&&(q.innerHTML=gt),tt=s(t),p(R.$$.fragment,t),et=s(t),A=d(t,"TABLE",{"data-svelte-h":!0}),f(A)!=="svelte-xzva6e"&&(A.innerHTML=xt),it=s(t),p(D.$$.fragment,t),at=s(t),I=d(t,"TABLE",{"data-svelte-h":!0}),f(I)!=="svelte-zeg51l"&&(I.innerHTML=ht),nt=s(t),p(L.$$.fragment,t),rt=s(t),N=d(t,"P",{}),kt(N).forEach(i),this.h()},h(){Lt(n,"name","hf:doc:metadata"),Lt(n,"content",qt)},m(t,e){Et(document.head,n),a(t,c,e),a(t,r,e),a(t,m,e),g(v,t,e),a(t,F,e),g(b,t,e),a(t,G,e),g(_,t,e),a(t,V,e),a(t,M,e),a(t,X,e),a(t,C,e),a(t,Q,e),a(t,E,e),a(t,U,e),a(t,P,e),a(t,W,e),a(t,H,e),a(t,j,e),g(w,t,e),a(t,K,e),g(S,t,e),a(t,O,e),a(t,y,e),a(t,Y,e),g(k,t,e),a(t,J,e),g(B,t,e),a(t,Z,e),a(t,q,e),a(t,tt,e),g(R,t,e),a(t,et,e),a(t,A,e),a(t,it,e),g(D,t,e),a(t,at,e),a(t,I,e),a(t,nt,e),g(L,t,e),a(t,rt,e),a(t,N,e),ot=!0},p(t,[e]){const $t={};e&2&&($t.$$scope={dirty:e,ctx:t}),w.$set($t);const Tt={};e&2&&(Tt.$$scope={dirty:e,ctx:t}),k.$set(Tt);const wt={};e&2&&(wt.$$scope={dirty:e,ctx:t}),L.$set(wt)},i(t){ot||(x(v.$$.fragment,t),x(b.$$.fragment,t),x(_.$$.fragment,t),x(w.$$.fragment,t),x(S.$$.fragment,t),x(k.$$.fragment,t),x(B.$$.fragment,t),x(R.$$.fragment,t),x(D.$$.fragment,t),x(L.$$.fragment,t),ot=!0)},o(t){h(v.$$.fragment,t),h(b.$$.fragment,t),h(_.$$.fragment,t),h(w.$$.fragment,t),h(S.$$.fragment,t),h(k.$$.fragment,t),h(B.$$.fragment,t),h(R.$$.fragment,t),h(D.$$.fragment,t),h(L.$$.fragment,t),ot=!1},d(t){t&&(i(c),i(r),i(m),i(F),i(G),i(V),i(M),i(X),i(C),i(Q),i(E),i(U),i(P),i(W),i(H),i(j),i(K),i(O),i(y),i(Y),i(J),i(Z),i(q),i(tt),i(et),i(A),i(it),i(at),i(I),i(nt),i(rt),i(N)),i(n),$(v,t),$(b,t),$(_,t),$(w,t),$(S,t),$(k,t),$(B,t),$(R,t),$(D,t),$(L,t)}}}const qt='{"title":"Supported architectures","local":"supported-architectures","sections":[{"title":"Training","local":"training","sections":[],"depth":2},{"title":"Inference","local":"inference","sections":[{"title":"Transformers","local":"transformers","sections":[],"depth":3},{"title":"Diffusers","local":"diffusers","sections":[],"depth":3},{"title":"Sentence Transformers","local":"sentence-transformers","sections":[],"depth":3}],"depth":2}],"depth":1}';function Rt(T){return bt(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class Ft extends _t{constructor(n){super(),Mt(this,n,Rt,Bt,vt,{})}}export{Ft as component};