14 lines
23 KiB
HTML
14 lines
23 KiB
HTML
<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><meta name="generator" content="rustdoc"><meta name="description" content="Candle implementations for various deep learning models"><title>candle_transformers::models - Rust</title><script>if(window.location.protocol!=="file:")document.head.insertAdjacentHTML("beforeend","SourceSerif4-Regular-6b053e98.ttf.woff2,FiraSans-Italic-81dc35de.woff2,FiraSans-Regular-0fe48ade.woff2,FiraSans-MediumItalic-ccf7e434.woff2,FiraSans-Medium-e1aa3f0a.woff2,SourceCodePro-Regular-8badfe75.ttf.woff2,SourceCodePro-Semibold-aa29a496.ttf.woff2".split(",").map(f=>`<link rel="preload" as="font" type="font/woff2"href="../../static.files/${f}">`).join(""))</script><link rel="stylesheet" href="../../static.files/normalize-9960930a.css"><link rel="stylesheet" href="../../static.files/rustdoc-77263533.css"><meta name="rustdoc-vars" data-root-path="../../" data-static-root-path="../../static.files/" data-current-crate="candle_transformers" data-themes="" data-resource-suffix="" data-rustdoc-version="1.94.1 (e408947bf 2026-03-25)" data-channel="1.94.1" data-search-js="search-9e2438ea.js" data-stringdex-js="stringdex-b897f86f.js" data-settings-js="settings-c38705f0.js" ><script src="../../static.files/storage-e2aeef58.js"></script><script defer src="../sidebar-items.js"></script><script defer src="../../static.files/main-7bab91a1.js"></script><noscript><link rel="stylesheet" href="../../static.files/noscript-ffcac47a.css"></noscript><link rel="alternate icon" type="image/png" href="../../static.files/favicon-32x32-eab170b8.png"><link rel="icon" type="image/svg+xml" href="../../static.files/favicon-044be391.svg"></head><body class="rustdoc mod"><!--[if lte IE 11]><div class="warning">This old browser is unsupported and will most likely display funky things.</div><![endif]--><rustdoc-topbar><h2><a href="#">Module models</a></h2></rustdoc-topbar><nav class="sidebar"><div class="sidebar-crate"><h2><a href="../../candle_transformers/index.html">candle_<wbr>transformers</a><span class="version">0.8.4</span></h2></div><div class="sidebar-elems"><section id="rustdoc-toc"><h2 class="location"><a href="#">Module models</a></h2><h3><a href="#modules">Module Items</a></h3><ul class="block"><li><a href="#modules" title="Modules">Modules</a></li></ul></section><div id="rustdoc-modnav"><h2 class="in-crate"><a href="../index.html">In crate candle_<wbr>transformers</a></h2></div></div></nav><div class="sidebar-resizer" title="Drag to resize sidebar"></div><main><div class="width-limiter"><section id="main-content" class="content"><div class="main-heading"><div class="rustdoc-breadcrumbs"><a href="../index.html">candle_transformers</a></div><h1>Module <span>models</span> <button id="copy-path" title="Copy item path to clipboard">Copy item path</button></h1><rustdoc-toolbar></rustdoc-toolbar><span class="sub-heading"><a class="src" href="../../src/candle_transformers/models/mod.rs.html#1-118">Source</a> </span></div><details class="toggle top-doc" open><summary class="hideme"><span>Expand description</span></summary><div class="docblock"><p>Candle implementations for various deep learning models</p>
|
||
<p>This crate provides implementations of popular machine learning models and architectures for different modalities.</p>
|
||
<ul>
|
||
<li>Large language models: <a href="llama/index.html" title="mod candle_transformers::models::llama"><code>llama</code></a>, <a href="phi3/index.html" title="mod candle_transformers::models::phi3"><code>phi3</code></a>, <a href="mamba/index.html" title="mod candle_transformers::models::mamba"><code>mamba</code></a>, <a href="mixtral/index.html" title="mod candle_transformers::models::mixtral"><code>mixtral</code></a>, <a href="bert/index.html" title="mod candle_transformers::models::bert"><code>bert</code></a>, …</li>
|
||
<li>Text to text models: <a href="t5/index.html" title="mod candle_transformers::models::t5"><code>t5</code></a>, …</li>
|
||
<li>Image to text models: <a href="blip/index.html" title="mod candle_transformers::models::blip"><code>blip</code></a>, …</li>
|
||
<li>Text to image models: <a href="stable_diffusion/index.html" title="mod candle_transformers::models::stable_diffusion"><code>stable_diffusion</code></a> and <a href="wuerstchen/index.html" title="mod candle_transformers::models::wuerstchen"><code>wuerstchen</code></a>, …</li>
|
||
<li>Audio models: <a href="whisper/index.html" title="mod candle_transformers::models::whisper"><code>whisper</code></a>, <a href="encodec/index.html" title="mod candle_transformers::models::encodec"><code>encodec</code></a>, <a href="metavoice/index.html" title="mod candle_transformers::models::metavoice"><code>metavoice</code></a>, <a href="parler_tts/index.html" title="mod candle_transformers::models::parler_tts"><code>parler_tts</code></a>, …</li>
|
||
<li>Computer vision models: <a href="dinov2/index.html" title="mod candle_transformers::models::dinov2"><code>dinov2</code></a>, <a href="convmixer/index.html" title="mod candle_transformers::models::convmixer"><code>convmixer</code></a>, <a href="efficientnet/index.html" title="mod candle_transformers::models::efficientnet"><code>efficientnet</code></a>, …</li>
|
||
</ul>
|
||
<p>Some of the models also have quantized variants, e.g. <a href="quantized_blip/index.html" title="mod candle_transformers::models::quantized_blip"><code>quantized_blip</code></a>, <a href="quantized_llama/index.html" title="mod candle_transformers::models::quantized_llama"><code>quantized_llama</code></a> and <a href="quantized_qwen2/index.html" title="mod candle_transformers::models::quantized_qwen2"><code>quantized_qwen2</code></a>.</p>
|
||
<p>The implementations aim to be readable while maintaining good performance. For more information
|
||
on each model see the model’s module docs in the links below.</p>
|
||
</div></details><h2 id="modules" class="section-header">Modules<a href="#modules" class="anchor">§</a></h2><dl class="item-table"><dt><a class="mod" href="based/index.html" title="mod candle_transformers::models::based">based</a></dt><dd>Based from the Stanford Hazy Research group.</dd><dt><a class="mod" href="beit/index.html" title="mod candle_transformers::models::beit">beit</a></dt><dd>Based on the BEIT vision-language model.</dd><dt><a class="mod" href="bert/index.html" title="mod candle_transformers::models::bert">bert</a></dt><dd>BERT (Bidirectional Encoder Representations from Transformers)</dd><dt><a class="mod" href="bigcode/index.html" title="mod candle_transformers::models::bigcode">bigcode</a></dt><dd>BigCode implementation in Rust based on the GPT-BigCode model.</dd><dt><a class="mod" href="blip/index.html" title="mod candle_transformers::models::blip">blip</a></dt><dd>Based on the BLIP paper from Salesforce Research.</dd><dt><a class="mod" href="blip_text/index.html" title="mod candle_transformers::models::blip_text">blip_<wbr>text</a></dt><dd>Implementation of BLIP text encoder/decoder.</dd><dt><a class="mod" href="chatglm/index.html" title="mod candle_transformers::models::chatglm">chatglm</a></dt><dd>Implementation of the ChatGLM2/3 models from THUDM.</dd><dt><a class="mod" href="chinese_clip/index.html" title="mod candle_transformers::models::chinese_clip">chinese_<wbr>clip</a></dt><dd>Chinese contrastive Language-Image Pre-Training</dd><dt><a class="mod" href="clip/index.html" title="mod candle_transformers::models::clip">clip</a></dt><dd>Contrastive Language-Image Pre-Training</dd><dt><a class="mod" href="codegeex4_9b/index.html" title="mod candle_transformers::models::codegeex4_9b">codegeex4_<wbr>9b</a></dt><dd>CodeGeeX4 - A multi-language code generation model</dd><dt><a class="mod" href="colpali/index.html" title="mod candle_transformers::models::colpali">colpali</a></dt><dd>Colpali Model for text/image similarity scoring.</dd><dt><a class="mod" href="convmixer/index.html" title="mod candle_transformers::models::convmixer">convmixer</a></dt><dd>ConvMixer implementation.</dd><dt><a class="mod" href="convnext/index.html" title="mod candle_transformers::models::convnext">convnext</a></dt><dd>ConvNeXt implementation.</dd><dt><a class="mod" href="dac/index.html" title="mod candle_transformers::models::dac">dac</a></dt><dd>Implementation of the Descript Audio Codec (DAC) model</dd><dt><a class="mod" href="debertav2/index.html" title="mod candle_transformers::models::debertav2">debertav2</a></dt><dt><a class="mod" href="deepseek2/index.html" title="mod candle_transformers::models::deepseek2">deepseek2</a></dt><dt><a class="mod" href="depth_anything_v2/index.html" title="mod candle_transformers::models::depth_anything_v2">depth_<wbr>anything_<wbr>v2</a></dt><dd>Implementation of the Depth Anything model from FAIR.</dd><dt><a class="mod" href="dinov2/index.html" title="mod candle_transformers::models::dinov2">dinov2</a></dt><dd>Implementation of the DINOv2 models from Meta Research.</dd><dt><a class="mod" href="dinov2reg4/index.html" title="mod candle_transformers::models::dinov2reg4">dinov2reg4</a></dt><dd>Implementation of the DINOv2 revision (4 regularization)</dd><dt><a class="mod" href="distilbert/index.html" title="mod candle_transformers::models::distilbert">distilbert</a></dt><dd>Implementation of DistilBert, a distilled version of BERT.</dd><dt><a class="mod" href="efficientnet/index.html" title="mod candle_transformers::models::efficientnet">efficientnet</a></dt><dd>Implementation of EfficientBert, an efficient variant of BERT for computer vision tasks.</dd><dt><a class="mod" href="efficientvit/index.html" title="mod candle_transformers::models::efficientvit">efficientvit</a></dt><dd>EfficientViT (MSRA) inference implementation based on timm.</dd><dt><a class="mod" href="encodec/index.html" title="mod candle_transformers::models::encodec">encodec</a></dt><dd>EnCodec neural audio codec based on the Encodec implementation.</dd><dt><a class="mod" href="eva2/index.html" title="mod candle_transformers::models::eva2">eva2</a></dt><dd>EVA-2 inference implementation.</dd><dt><a class="mod" href="falcon/index.html" title="mod candle_transformers::models::falcon">falcon</a></dt><dd>Falcon language model inference implementation</dd><dt><a class="mod" href="fastvit/index.html" title="mod candle_transformers::models::fastvit">fastvit</a></dt><dd>FastViT inference implementation based on timm</dd><dt><a class="mod" href="flux/index.html" title="mod candle_transformers::models::flux">flux</a></dt><dd>Flux Model</dd><dt><a class="mod" href="gemma/index.html" title="mod candle_transformers::models::gemma">gemma</a></dt><dd>Gemma inference implementation.</dd><dt><a class="mod" href="gemma2/index.html" title="mod candle_transformers::models::gemma2">gemma2</a></dt><dd>Gemma LLM architecture (Google) inference implementation.</dd><dt><a class="mod" href="gemma3/index.html" title="mod candle_transformers::models::gemma3">gemma3</a></dt><dd>Gemma LLM architecture (Google) inference implementation.</dd><dt><a class="mod" href="glm4/index.html" title="mod candle_transformers::models::glm4">glm4</a></dt><dd>GLM-4 inference implementation.</dd><dt><a class="mod" href="granite/index.html" title="mod candle_transformers::models::granite">granite</a></dt><dd>Granite is a Long Context Transformer Language Model.</dd><dt><a class="mod" href="helium/index.html" title="mod candle_transformers::models::helium">helium</a></dt><dd>Helium inference implementation.</dd><dt><a class="mod" href="hiera/index.html" title="mod candle_transformers::models::hiera">hiera</a></dt><dd>Hiera inference implementation based on timm.</dd><dt><a class="mod" href="jina_bert/index.html" title="mod candle_transformers::models::jina_bert">jina_<wbr>bert</a></dt><dd>JinaBERT inference implementation</dd><dt><a class="mod" href="llama/index.html" title="mod candle_transformers::models::llama">llama</a></dt><dd>Llama inference implementation.</dd><dt><a class="mod" href="llama2_c/index.html" title="mod candle_transformers::models::llama2_c">llama2_<wbr>c</a></dt><dd>Llama2 inference implementation.</dd><dt><a class="mod" href="llama2_c_weights/index.html" title="mod candle_transformers::models::llama2_c_weights">llama2_<wbr>c_<wbr>weights</a></dt><dd>Llama2 inference implementation.</dd><dt><a class="mod" href="llava/index.html" title="mod candle_transformers::models::llava">llava</a></dt><dd>The LLaVA (Large Language and Vision Assistant) model.</dd><dt><a class="mod" href="mamba/index.html" title="mod candle_transformers::models::mamba">mamba</a></dt><dd>Mamba inference implementation.</dd><dt><a class="mod" href="marian/index.html" title="mod candle_transformers::models::marian">marian</a></dt><dd>Marian Neural Machine Translation</dd><dt><a class="mod" href="metavoice/index.html" title="mod candle_transformers::models::metavoice">metavoice</a></dt><dd>MetaVoice Studio ML Models</dd><dt><a class="mod" href="mimi/index.html" title="mod candle_transformers::models::mimi">mimi</a></dt><dd>mimi model</dd><dt><a class="mod" href="mistral/index.html" title="mod candle_transformers::models::mistral">mistral</a></dt><dd>Mixtral Model, based on the Mistral architecture</dd><dt><a class="mod" href="mixformer/index.html" title="mod candle_transformers::models::mixformer">mixformer</a></dt><dd>MixFormer (Microsoft’s Phi Architecture)</dd><dt><a class="mod" href="mixtral/index.html" title="mod candle_transformers::models::mixtral">mixtral</a></dt><dd>Mixtral Model, a sparse mixture of expert model based on the Mistral architecture</dd><dt><a class="mod" href="mmdit/index.html" title="mod candle_transformers::models::mmdit">mmdit</a></dt><dd>Mix of Multi-scale Dilated and Traditional Convolutions</dd><dt><a class="mod" href="mobileclip/index.html" title="mod candle_transformers::models::mobileclip">mobileclip</a></dt><dd>Mobile CLIP model, combining a lightweight vision encoder with a text encoder</dd><dt><a class="mod" href="mobilenetv4/index.html" title="mod candle_transformers::models::mobilenetv4">mobilenetv4</a></dt><dd>MobileNet-v4</dd><dt><a class="mod" href="mobileone/index.html" title="mod candle_transformers::models::mobileone">mobileone</a></dt><dd>MobileOne</dd><dt><a class="mod" href="modernbert/index.html" title="mod candle_transformers::models::modernbert">modernbert</a></dt><dd>ModernBERT</dd><dt><a class="mod" href="moondream/index.html" title="mod candle_transformers::models::moondream">moondream</a></dt><dd>MoonDream Model vision-to-text</dd><dt><a class="mod" href="mpt/index.html" title="mod candle_transformers::models::mpt">mpt</a></dt><dd>Module implementing the MPT (Multi-Purpose Transformer) model</dd><dt><a class="mod" href="nvembed_v2/index.html" title="mod candle_transformers::models::nvembed_v2">nvembed_<wbr>v2</a></dt><dd>NV-Embed-v2</dd><dt><a class="mod" href="olmo/index.html" title="mod candle_transformers::models::olmo">olmo</a></dt><dd>OLMo (Open Language Model) implementation</dd><dt><a class="mod" href="openclip/index.html" title="mod candle_transformers::models::openclip">openclip</a></dt><dd>Open Contrastive Language-Image Pre-Training</dd><dt><a class="mod" href="paligemma/index.html" title="mod candle_transformers::models::paligemma">paligemma</a></dt><dd>Multimodal multi-purpose model combining Gemma-based language model with SigLIP image understanding</dd><dt><a class="mod" href="parler_tts/index.html" title="mod candle_transformers::models::parler_tts">parler_<wbr>tts</a></dt><dd>Parler Model implementation for parler_tts text-to-speech synthesis</dd><dt><a class="mod" href="persimmon/index.html" title="mod candle_transformers::models::persimmon">persimmon</a></dt><dd>Persimmon Model</dd><dt><a class="mod" href="phi/index.html" title="mod candle_transformers::models::phi">phi</a></dt><dd>Microsoft Phi model implementation</dd><dt><a class="mod" href="phi3/index.html" title="mod candle_transformers::models::phi3">phi3</a></dt><dd>Microsoft Phi-3 model implementation</dd><dt><a class="mod" href="pixtral/index.html" title="mod candle_transformers::models::pixtral">pixtral</a></dt><dd>Pixtral Language-Image Pre-Training</dd><dt><a class="mod" href="quantized_blip/index.html" title="mod candle_transformers::models::quantized_blip">quantized_<wbr>blip</a></dt><dd>BLIP model implementation with quantization support.</dd><dt><a class="mod" href="quantized_blip_text/index.html" title="mod candle_transformers::models::quantized_blip_text">quantized_<wbr>blip_<wbr>text</a></dt><dd>Quantized BLIP text module implementation.</dd><dt><a class="mod" href="quantized_llama/index.html" title="mod candle_transformers::models::quantized_llama">quantized_<wbr>llama</a></dt><dd>Quantized llama model implementation.</dd><dt><a class="mod" href="quantized_llama2_c/index.html" title="mod candle_transformers::models::quantized_llama2_c">quantized_<wbr>llama2_<wbr>c</a></dt><dd>Quantized Llama2 model implementation.</dd><dt><a class="mod" href="quantized_metavoice/index.html" title="mod candle_transformers::models::quantized_metavoice">quantized_<wbr>metavoice</a></dt><dd>Quantized MetaVoice model implementation.</dd><dt><a class="mod" href="quantized_mistral/index.html" title="mod candle_transformers::models::quantized_mistral">quantized_<wbr>mistral</a></dt><dd>Mistral model implementation with quantization support.</dd><dt><a class="mod" href="quantized_mixformer/index.html" title="mod candle_transformers::models::quantized_mixformer">quantized_<wbr>mixformer</a></dt><dd>Module containing quantized MixFormer model implementation.</dd><dt><a class="mod" href="quantized_moondream/index.html" title="mod candle_transformers::models::quantized_moondream">quantized_<wbr>moondream</a></dt><dd>Implementation of a quantized Moondream vision language model.</dd><dt><a class="mod" href="quantized_mpt/index.html" title="mod candle_transformers::models::quantized_mpt">quantized_<wbr>mpt</a></dt><dd>Quantized MPT model implementation.</dd><dt><a class="mod" href="quantized_phi/index.html" title="mod candle_transformers::models::quantized_phi">quantized_<wbr>phi</a></dt><dd>Phi2 model implementation with quantization support.</dd><dt><a class="mod" href="quantized_phi3/index.html" title="mod candle_transformers::models::quantized_phi3">quantized_<wbr>phi3</a></dt><dd>Phi3 model implementation with quantization support.</dd><dt><a class="mod" href="quantized_qwen2/index.html" title="mod candle_transformers::models::quantized_qwen2">quantized_<wbr>qwen2</a></dt><dd>Qwen2 model implementation with quantization support.</dd><dt><a class="mod" href="quantized_recurrent_gemma/index.html" title="mod candle_transformers::models::quantized_recurrent_gemma">quantized_<wbr>recurrent_<wbr>gemma</a></dt><dd>Recurrent Gemma model implementation with quantization support.</dd><dt><a class="mod" href="quantized_rwkv_v5/index.html" title="mod candle_transformers::models::quantized_rwkv_v5">quantized_<wbr>rwkv_<wbr>v5</a></dt><dd>RWKV v5 model implementation with quantization support.</dd><dt><a class="mod" href="quantized_rwkv_v6/index.html" title="mod candle_transformers::models::quantized_rwkv_v6">quantized_<wbr>rwkv_<wbr>v6</a></dt><dd>RWKV v6 model implementation with quantization support.</dd><dt><a class="mod" href="quantized_stable_lm/index.html" title="mod candle_transformers::models::quantized_stable_lm">quantized_<wbr>stable_<wbr>lm</a></dt><dd>Module for quantized StableLM implementation.</dd><dt><a class="mod" href="quantized_t5/index.html" title="mod candle_transformers::models::quantized_t5">quantized_<wbr>t5</a></dt><dd>T5 model implementation with quantization support.</dd><dt><a class="mod" href="qwen2/index.html" title="mod candle_transformers::models::qwen2">qwen2</a></dt><dd>Qwen2 model implementation with quantization support.</dd><dt><a class="mod" href="qwen2_moe/index.html" title="mod candle_transformers::models::qwen2_moe">qwen2_<wbr>moe</a></dt><dd>Qwen2 model implementation with Mixture of Experts support.</dd><dt><a class="mod" href="recurrent_gemma/index.html" title="mod candle_transformers::models::recurrent_gemma">recurrent_<wbr>gemma</a></dt><dd>Recurrent Gemma model implementation</dd><dt><a class="mod" href="repvgg/index.html" title="mod candle_transformers::models::repvgg">repvgg</a></dt><dd>RepVGG inference implementation</dd><dt><a class="mod" href="resnet/index.html" title="mod candle_transformers::models::resnet">resnet</a></dt><dd>ResNet Implementation</dd><dt><a class="mod" href="rwkv_v5/index.html" title="mod candle_transformers::models::rwkv_v5">rwkv_v5</a></dt><dd>RWKV v5 model implementation.</dd><dt><a class="mod" href="rwkv_v6/index.html" title="mod candle_transformers::models::rwkv_v6">rwkv_v6</a></dt><dd>RWKV v6 model implementation.</dd><dt><a class="mod" href="segformer/index.html" title="mod candle_transformers::models::segformer">segformer</a></dt><dd>Segformer model implementation for semantic segmentation and image classification.</dd><dt><a class="mod" href="segment_anything/index.html" title="mod candle_transformers::models::segment_anything">segment_<wbr>anything</a></dt><dd>Segment Anything Model (SAM)</dd><dt><a class="mod" href="siglip/index.html" title="mod candle_transformers::models::siglip">siglip</a></dt><dd>Siglip model implementation.</dd><dt><a class="mod" href="stable_diffusion/index.html" title="mod candle_transformers::models::stable_diffusion">stable_<wbr>diffusion</a></dt><dd>Stable Diffusion</dd><dt><a class="mod" href="stable_lm/index.html" title="mod candle_transformers::models::stable_lm">stable_<wbr>lm</a></dt><dd>StableLM model implementation.</dd><dt><a class="mod" href="starcoder2/index.html" title="mod candle_transformers::models::starcoder2">starcoder2</a></dt><dd>StarCoder model implementation with quantization support.</dd><dt><a class="mod" href="stella_en_v5/index.html" title="mod candle_transformers::models::stella_en_v5">stella_<wbr>en_<wbr>v5</a></dt><dd>Stella v5 model implementation.</dd><dt><a class="mod" href="t5/index.html" title="mod candle_transformers::models::t5">t5</a></dt><dd>T5 model implementation.</dd><dt><a class="mod" href="trocr/index.html" title="mod candle_transformers::models::trocr">trocr</a></dt><dd>TrOCR model implementation.</dd><dt><a class="mod" href="vgg/index.html" title="mod candle_transformers::models::vgg">vgg</a></dt><dd>VGG-16 model implementation.</dd><dt><a class="mod" href="vit/index.html" title="mod candle_transformers::models::vit">vit</a></dt><dd>Vision Transformer (ViT) implementation.</dd><dt><a class="mod" href="whisper/index.html" title="mod candle_transformers::models::whisper">whisper</a></dt><dd>Whisper Model Implementation</dd><dt><a class="mod" href="with_tracing/index.html" title="mod candle_transformers::models::with_tracing">with_<wbr>tracing</a></dt><dt><a class="mod" href="wuerstchen/index.html" title="mod candle_transformers::models::wuerstchen">wuerstchen</a></dt><dd>Würstchen Efficient Diffusion Model</dd><dt><a class="mod" href="xlm_roberta/index.html" title="mod candle_transformers::models::xlm_roberta">xlm_<wbr>roberta</a></dt><dt><a class="mod" href="yi/index.html" title="mod candle_transformers::models::yi">yi</a></dt><dd>Yi model implementation.</dd></dl></section></div></main></body></html> |