{"id":10943,"date":"2024-09-30T00:54:42","date_gmt":"2024-09-29T15:54:42","guid":{"rendered":"https:\/\/www.stechstar.com\/user\/wordpress\/?p=10943"},"modified":"2026-08-17T09:37:34","modified_gmt":"2026-08-17T00:37:34","slug":"%ec%9d%b8%ea%b3%b5%ec%a7%80%eb%8a%a5-%ea%b8%b0%ec%88%a0-%ec%86%8c%ea%b7%9c%eb%aa%a8-%ec%96%b8%ec%96%b4-%eb%aa%a8%eb%8d%b8slm-%ec%86%8c%ea%b7%9c%eb%aa%a8-%ec%96%b8%ec%96%b4-%eb%aa%a8%eb%8d%b8","status":"publish","type":"post","link":"https:\/\/www.stechstar.com\/user\/wordpress\/%ec%9d%b8%ea%b3%b5%ec%a7%80%eb%8a%a5-%ea%b8%b0%ec%88%a0-%ec%86%8c%ea%b7%9c%eb%aa%a8-%ec%96%b8%ec%96%b4-%eb%aa%a8%eb%8d%b8slm-%ec%86%8c%ea%b7%9c%eb%aa%a8-%ec%96%b8%ec%96%b4-%eb%aa%a8%eb%8d%b8\/","title":{"rendered":"[\uc778\uacf5\uc9c0\ub2a5 \uae30\uc220] \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM) \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc758 \ubd80\uc0c1: AI\ub97c \uc704\ud55c \ud6a8\uc728\uc131\uacfc \ub9de\ucda4\ud654 : Small Language Models (SLMs)"},"content":{"rendered":"<h1><span style=\"color: #000000;\">[\uc778\uacf5\uc9c0\ub2a5 \uae30\uc220] \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM) \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc758 \ubd80\uc0c1: AI\ub97c \uc704\ud55c \ud6a8\uc728\uc131\uacfc \ub9de\ucda4\ud654 : Small Language Models (SLMs)<\/span><\/h1>\n<div>\n<h1 id=\"94fc\" class=\"pw-post-title fo fp fq bf fr fs ft fu fv fw fx fy fz ga gb gc gd ge gf gg gh gi gj gk gl gm bk\" data-selectable-paragraph=\"\" data-testid=\"storyTitle\"><span style=\"color: #000000;\">Small Language Models (SLMs)<\/span><\/h1>\n<\/div>\n<div>\n<h2 id=\"6abb\" class=\"pw-subtitle-paragraph gn fp fq bf b go gp gq gr gs gt gu gv gw gx gy gz ha hb hc cq dx\" data-selectable-paragraph=\"\"><span class=\"al\" style=\"color: #000000;\">The Rise of Small Language Models: Efficiency and Customization for AI<\/span><\/h2>\n<div>\n<div class=\"speechify-ignore ab cp\">\n<div class=\"speechify-ignore bh l\">\n<div class=\"hd he hf hg hh ab\">\n<div>\n<div class=\"ab hi\">\n<div>\n<div class=\"bm\" aria-describedby=\"1\" aria-hidden=\"false\" aria-labelledby=\"1\">\n<div class=\"l hj hk by hl hm\">\n<div class=\"l ed\">\n<p><span style=\"color: #000000;\"><a style=\"color: #000000;\" href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\" rel=\"noopener follow\" data-cke-saved-href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\"><img loading=\"lazy\" decoding=\"async\" class=\"l ep by dd de cx\" src=\"https:\/\/stechstar.com\/user\/zbxe\/files\/attach\/images\/92879\/020\/104\/0a148b999ace16a52a29af294aee8a4d.jpeg\" alt=\"Nagesh Mashette\" width=\"44\" height=\"44\" data-autoattach=\"success\" data-testid=\"authorPhoto\" data-cke-saved-src=\".\/files\/attach\/images\/92879\/020\/104\/0a148b999ace16a52a29af294aee8a4d.jpeg\"><\/a><\/span><\/p>\n<div class=\"hn by l dd de em n ho eo\"><a class=\"af ag ah ai aj ak al am an ao ap aq ar ht\" style=\"color: #000000;\" href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\" rel=\"noopener follow\" data-testid=\"authorName\" data-cke-saved-href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\">Nagesh Mashette<\/a><\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<div class=\"l ib\">\n<figure class=\"lv lw lx ly lz ma ls lt paragraph-image\">\n<div class=\"mb mc ed md bh me\" tabindex=\"0\" role=\"button\">\n<div class=\"ls lt lu\"><span style=\"color: #000000;\"><picture><img loading=\"lazy\" decoding=\"async\" class=\"bh la mf c\" role=\"presentation\" src=\"https:\/\/stechstar.com\/user\/zbxe\/files\/attach\/images\/92879\/020\/104\/858a1507b485449719500a56051cbbcb.png\" alt=\"\" width=\"700\" height=\"314\" data-autoattach=\"success\" data-cke-saved-src=\".\/files\/attach\/images\/92879\/020\/104\/858a1507b485449719500a56051cbbcb.png\"><\/picture><\/span><\/div>\n<\/div>\n<\/figure>\n<p id=\"42ad\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Large language models (LLMs) have captured headlines and imaginations with their impressive capabilities in natural language processing. However, their massive size and resource requirements have limited their accessibility and applicability. Enter the small language model (SLM), a compact and efficient alternative poised to democratize AI for diverse needs.<\/span><\/p>\n<h1 id=\"8445\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">What are Small Language Models?<\/span><\/h1>\n<p id=\"168b\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">SLMs are essentially smaller versions of their LLM counterparts. They have significantly fewer parameters, typically ranging from a few million to a few billion, compared to LLMs with hundreds of billions or even trillions. This difference in size translates to several advantages:<\/span><\/p>\n<ul>\n<li id=\"cabc\" class=\"mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Efficiency: SLMs require less computational power and memory, making them suitable for deployment on smaller devices or even edge computing scenarios. This opens up opportunities for real-world applications like on-device chatbots and personalized mobile assistants.<\/span><\/li>\n<li id=\"ad86\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Accessibility: With lower resource requirements, SLMs are more accessible to a broader range of developers and organizations. This democratizes AI, allowing smaller teams and individual researchers to explore the power of language models without significant infrastructure investments.<\/span><\/li>\n<li id=\"3f9f\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Customization: SLMs are easier to fine-tune for specific domains and tasks. This enables the creation of specialized models tailored to niche applications, leading to higher performance and accuracy.<\/span><\/li>\n<\/ul>\n<h1 id=\"229c\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">How do Small Language Models Work?<\/span><\/h1>\n<p id=\"5c44\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Like LLMs, SLMs are trained on massive datasets of text and code. However, several techniques are employed to achieve their smaller size and efficiency:<\/span><\/p>\n<ul>\n<li id=\"e90e\" class=\"mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Knowledge Distillation: This involves transferring knowledge from a pre-trained LLM to a smaller model, capturing its core capabilities without the full complexity.<\/span><\/li>\n<li id=\"b0f9\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Pruning and Quantization: These techniques remove unnecessary parts of the model and reduce the precision of its weights, respectively, further reducing its size and resource requirements.<\/span><\/li>\n<li id=\"0c7f\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Efficient Architectures: Researchers are continually developing novel architectures specifically designed for SLMs, focusing on optimizing both performance and efficiency.<\/span><\/li>\n<\/ul>\n<h1 id=\"8ebc\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Benefits and Limitations<\/span><\/h1>\n<p id=\"cf6e\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Small Language Models (SLMs) offer the advantage of being trainable with relatively modest datasets. Their simplified architectures enhance interpretability, and their compact size facilitates deployment on mobile devices.<\/span><\/p>\n<p id=\"0699\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">A notable benefit of SLMs is their capability to process data locally, making them particularly valuable for Internet of Things (IoT) edge devices and enterprises bound by stringent privacy and security regulations.<\/span><\/p>\n<p id=\"da95\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">However, deploying small language models involves a trade-off. Due to their training on smaller datasets, SLMs possess more constrained knowledge bases compared to their Large Language Model (LLM) counterparts. Additionally, their understanding of language and context tends to be more limited, potentially resulting in less accurate and nuanced responses when compared to larger models.<\/span><\/p>\n<figure class=\"lv lw lx ly lz ma ls lt paragraph-image\">\n<div class=\"mb mc ed md bh me\" tabindex=\"0\" role=\"button\">\n<div class=\"ls lt ol\"><span style=\"color: #000000;\"><picture><source srcset=\"https:\/\/miro.medium.com\/v2\/resize:fit:640\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 640w, https:\/\/miro.medium.com\/v2\/resize:fit:720\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 720w, https:\/\/miro.medium.com\/v2\/resize:fit:750\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 750w, https:\/\/miro.medium.com\/v2\/resize:fit:786\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 786w, https:\/\/miro.medium.com\/v2\/resize:fit:828\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 828w, https:\/\/miro.medium.com\/v2\/resize:fit:1100\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1100w, https:\/\/miro.medium.com\/v2\/resize:fit:1400\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1400w\" type=\"image\/webp\" sizes=\"(min-resolution: 4dppx) and (max-width: 700px) 50vw, (-webkit-min-device-pixel-ratio: 4) and (max-width: 700px) 50vw, (min-resolution: 3dppx) and (max-width: 700px) 67vw, (-webkit-min-device-pixel-ratio: 3) and (max-width: 700px) 65vw, (min-resolution: 2.5dppx) and (max-width: 700px) 80vw, (-webkit-min-device-pixel-ratio: 2.5) and (max-width: 700px) 80vw, (min-resolution: 2dppx) and (max-width: 700px) 100vw, (-webkit-min-device-pixel-ratio: 2) and (max-width: 700px) 100vw, 700px\"><source srcset=\"https:\/\/miro.medium.com\/v2\/resize:fit:640\/1*YyfhMNzNOcFzpF-YoRWy9w.png 640w, https:\/\/miro.medium.com\/v2\/resize:fit:720\/1*YyfhMNzNOcFzpF-YoRWy9w.png 720w, https:\/\/miro.medium.com\/v2\/resize:fit:750\/1*YyfhMNzNOcFzpF-YoRWy9w.png 750w, https:\/\/miro.medium.com\/v2\/resize:fit:786\/1*YyfhMNzNOcFzpF-YoRWy9w.png 786w, https:\/\/miro.medium.com\/v2\/resize:fit:828\/1*YyfhMNzNOcFzpF-YoRWy9w.png 828w, https:\/\/miro.medium.com\/v2\/resize:fit:1100\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1100w, https:\/\/miro.medium.com\/v2\/resize:fit:1400\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1400w\" sizes=\"(min-resolution: 4dppx) and (max-width: 700px) 50vw, (-webkit-min-device-pixel-ratio: 4) and (max-width: 700px) 50vw, (min-resolution: 3dppx) and (max-width: 700px) 67vw, (-webkit-min-device-pixel-ratio: 3) and (max-width: 700px) 65vw, (min-resolution: 2.5dppx) and (max-width: 700px) 80vw, (-webkit-min-device-pixel-ratio: 2.5) and (max-width: 700px) 80vw, (min-resolution: 2dppx) and (max-width: 700px) 100vw, (-webkit-min-device-pixel-ratio: 2) and (max-width: 700px) 100vw, 700px\" data-testid=\"og\"><img loading=\"lazy\" decoding=\"async\" class=\"bh la mf c\" role=\"presentation\" src=\"https:\/\/stechstar.com\/user\/zbxe\/files\/attach\/images\/92879\/020\/104\/de29769bbc3ce7b4fc50af0e1e6eddcd.png\" alt=\"\" width=\"700\" height=\"296\" data-autoattach=\"success\" data-cke-saved-src=\".\/files\/attach\/images\/92879\/020\/104\/de29769bbc3ce7b4fc50af0e1e6eddcd.png\"><\/picture><\/span><\/div>\n<\/div><figcaption class=\"om on oo ls lt op oq bf b bg z dx\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Comparision of SLM and LLM<\/span><\/figcaption><\/figure>\n<h1 id=\"47bf\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">Some Examples of Small Language Models (SLMs)<\/span><\/h1>\n<ol>\n<li id=\"2117\" class=\"mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">DistilBERT<\/span>: DistilBERT represents a more compact, agile, and lightweight iteration of BERT, a pioneering model in natural language processing (NLP). \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/distilbert\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/distilbert\">https:\/\/huggingface.co\/docs\/transformers\/model_doc\/distilbert<\/a><\/span><\/li>\n<li id=\"ed72\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">Orca 2<\/span>: Developed by Microsoft, Orca 2 is the result of fine-tuning Meta\u2019s Llama 2 using high-quality synthetic data. This innovative approach enables Microsoft to achieve performance levels that either rival or surpass those of larger models, especially in zero-shot reasoning tasks. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/microsoft\/Orca-2-13b\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/microsoft\/Orca-2-13b\">https:\/\/huggingface.co\/microsoft\/Orca-2-13b<\/a><\/span><\/li>\n<li id=\"e520\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">Phi 2<\/span>: Microsoft\u2019s Phi 2 is a transformer-based Small Language Model (SLM) engineered for efficiency and adaptability in both cloud and edge deployments. According to Microsoft, Phi 2 exhibits state-of-the-art performance in domains such as mathematical reasoning, common sense, language understanding, and logical reasoning. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/main\/model_doc\/phi\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/main\/model_doc\/phi\">https:\/\/huggingface.co\/docs\/transformers\/main\/model_doc\/phi<\/a><\/span><\/li>\n<li id=\"bbae\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">BERT Mini, Small, Medium, and Tiny<\/span>: Google\u2019s BERT model is available in scaled-down versions \u2014 ranging from Mini with 4.4 million parameters to Medium with 41 million parameters \u2014 to accommodate various resource constraints. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/prajjwal1\/bert-mini\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/prajjwal1\/bert-mini\">https:\/\/huggingface.co\/prajjwal1\/bert-mini<\/a><\/span><\/li>\n<li id=\"98e3\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">GPT-Neo and GPT-J<\/span>: GPT-Neo and GPT-J are scaled-down iterations of OpenAI\u2019s GPT models, offering versatility in application scenarios with more limited computational resources. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/gpt_neo\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/gpt_neo\">https:\/\/huggingface.co\/docs\/transformers\/model_doc\/gpt_neo<\/a><\/span><\/li>\n<li id=\"89e2\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">MobileBERT<\/span>: Tailored for mobile devices, MobileBERT is specifically designed to optimize performance within the constraints of mobile computing. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/mobilebert\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/mobilebert\">https:\/\/huggingface.co\/docs\/transformers\/model_doc\/mobilebert<\/a><\/span><\/li>\n<li id=\"d89c\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">T5-Small<\/span>: As part of Google\u2019s Text-to-Text Transfer Transformer (T5) model series, T5-Small strikes a balance between performance and resource utilization, aiming to provide efficient text processing capabilities. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/t5-small\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/t5-small\">https:\/\/huggingface.co\/t5-small<\/a><\/span><\/li>\n<\/ol>\n<h1 id=\"cc53\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">The Future of Small Language Models<\/span><\/h1>\n<p id=\"0ff5\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">As research and development progress, we can expect SLMs to become even more powerful and versatile. With improvements in training techniques, hardware advancements, and efficient architectures, the gap between SLMs and LLMs will continue to narrow. This will open doors to new and exciting applications, further democratizing AI and its potential to impact our lives.<\/span><\/p>\n<p id=\"f9c7\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">In conclusion, small language models represent a significant shift in the landscape of AI. Their efficiency, accessibility, and customization capabilities make them a valuable tool for developers and researchers across various domains. As SLMs continue to evolve, they hold immense promise to empower individuals and organizations alike, shaping a future where AI is not just powerful, but also accessible and tailored to diverse needs.<\/span><\/p>\n<p class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">[\ucd9c\ucc98]&nbsp;https:\/\/medium.com\/@nageshmashette32\/small-language-models-slms-305597c9edf2<\/span><\/p>\n<div>\n<h1 id=\"94fc\" class=\"pw-post-title fo fp fq bf fr fs ft fu fv fw fx fy fz ga gb gc gd ge gf gg gh gi gj gk gl gm bk\" data-selectable-paragraph=\"\" data-testid=\"storyTitle\"><span style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM)<\/span><\/h1>\n<\/div>\n<div>\n<h2 id=\"6abb\" class=\"pw-subtitle-paragraph gn fp fq bf b go gp gq gr gs gt gu gv gw gx gy gz ha hb hc cq dx\" data-selectable-paragraph=\"\"><span class=\"al\" style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc758 \ubd80\uc0c1: AI\ub97c \uc704\ud55c \ud6a8\uc728\uc131\uacfc \ub9de\ucda4\ud654<\/span><\/h2>\n<div>\n<div class=\"speechify-ignore ab cp\">\n<div class=\"speechify-ignore bh l\">\n<div class=\"hd he hf hg hh ab\">\n<div>\n<div class=\"ab hi\">\n<div>\n<div class=\"bm\" aria-describedby=\"1\" aria-hidden=\"false\" aria-labelledby=\"1\">\n<div class=\"l hj hk by hl hm\">\n<div class=\"l ed\">\n<p><span style=\"color: #000000;\"><a style=\"color: #000000;\" href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\" rel=\"noopener follow\" data-cke-saved-href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\"><img loading=\"lazy\" decoding=\"async\" class=\"l ep by dd de cx\" src=\"https:\/\/stechstar.com\/user\/zbxe\/files\/attach\/images\/92879\/020\/104\/0a148b999ace16a52a29af294aee8a4d.jpeg\" alt=\"\ub098\uac8c\uc26c \ub9c8\uc170\ud14c\" width=\"44\" height=\"44\" data-autoattach=\"success\" data-testid=\"authorPhoto\" data-cke-saved-src=\".\/files\/attach\/images\/92879\/020\/104\/0a148b999ace16a52a29af294aee8a4d.jpeg\"><\/a><\/span><\/p>\n<div class=\"hn by l dd de em n ho eo\"><a class=\"af ag ah ai aj ak al am an ao ap aq ar ht\" style=\"color: #000000;\" href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\" rel=\"noopener follow\" data-testid=\"authorName\" data-cke-saved-href=\"https:\/\/medium.com\/@nageshmashette32?source=post_page-----305597c9edf2--------------------------------\">\ub098\uac8c\uc26c \ub9c8\uc170\ud14c<\/a><\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<div class=\"bn bh l\">\n<div class=\"ab\">\n<div class=\"hp ab q\">\n<div class=\"ab q hq\">\n<p><span class=\"bf b bg z bk\" style=\"color: #000000;\"><span class=\"hu hv\" aria-hidden=\"true\"><span class=\"bf b bg z dx\">\u00b7<\/span><\/span><\/span><\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<figure class=\"lv lw lx ly lz ma ls lt paragraph-image\">\n<div class=\"mb mc ed md bh me\" tabindex=\"0\" role=\"button\">\n<div class=\"ls lt lu\"><span style=\"color: #000000;\"><picture><img loading=\"lazy\" decoding=\"async\" class=\"bh la mf c\" role=\"presentation\" src=\"https:\/\/stechstar.com\/user\/zbxe\/files\/attach\/images\/92879\/020\/104\/858a1507b485449719500a56051cbbcb.png\" alt=\"\" width=\"700\" height=\"314\" data-autoattach=\"success\" data-cke-saved-src=\".\/files\/attach\/images\/92879\/020\/104\/858a1507b485449719500a56051cbbcb.png\"><\/picture><\/span><\/div>\n<\/div>\n<\/figure>\n<p id=\"42ad\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\ub300\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(LLM)\uc740 \uc790\uc5f0\uc5b4 \ucc98\ub9ac\uc5d0\uc11c \uc778\uc0c1\uc801\uc778 \uc5ed\ub7c9\uc73c\ub85c \ud5e4\ub4dc\ub77c\uc778\uacfc \uc0c1\uc0c1\ub825\uc744 \uc0ac\ub85c\uc7a1\uc558\uc2b5\ub2c8\ub2e4. \uadf8\ub7ec\ub098 \uc5c4\uccad\ub09c \ud06c\uae30\uc640 \ub9ac\uc18c\uc2a4 \uc694\uad6c \uc0ac\ud56d\uc73c\ub85c \uc778\ud574 \uc811\uadfc\uc131\uacfc \uc801\uc6a9\uc131\uc774 \uc81c\ud55c\ub418\uc5c8\uc2b5\ub2c8\ub2e4. \ub2e4\uc591\ud55c \uc694\uad6c\uc5d0 \ub9de\uac8c AI\ub97c \ubbfc\uc8fc\ud654\ud560 \uc218 \uc788\ub294 \ucef4\ud329\ud2b8\ud558\uace0 \ud6a8\uc728\uc801\uc778 \ub300\uc548\uc778 \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM)\uc774 \ub4f1\uc7a5\ud588\uc2b5\ub2c8\ub2e4.<\/span><\/p>\n<h1 id=\"8445\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc774\ub780?<\/span><\/h1>\n<p id=\"168b\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">SLM\uc740 \ubcf8\uc9c8\uc801\uc73c\ub85c LLM \ub300\uc751\ubb3c\uc758 \ub354 \uc791\uc740 \ubc84\uc804\uc785\ub2c8\ub2e4. LLM\uc774 \uc218\ubc31\uc5b5 \ub610\ub294 \uc218\uc870 \uac1c\uc5d0 \ub2ec\ud558\ub294 \uac83\uacfc \ube44\uad50\ud588\uc744 \ub54c, \uc77c\ubc18\uc801\uc73c\ub85c \uc218\ubc31\ub9cc\uc5d0\uc11c \uc218\uc2ed\uc5b5 \uac1c\uc5d0 \uc774\ub974\ub294 \ub9e4\uac1c\ubcc0\uc218\uac00 \ud6e8\uc52c \uc801\uc2b5\ub2c8\ub2e4. \uc774\ub7ec\ud55c \ud06c\uae30\uc758 \ucc28\uc774\ub294 \uc5ec\ub7ec \uac00\uc9c0 \uc774\uc810\uc73c\ub85c \uc774\uc5b4\uc9d1\ub2c8\ub2e4.<\/span><\/p>\n<ul>\n<li id=\"cabc\" class=\"mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\ud6a8\uc728\uc131: SLM\uc740 \ucef4\ud4e8\ud305 \ud30c\uc6cc\uc640 \uba54\ubaa8\ub9ac\uac00 \ub35c \ud544\uc694\ud558\ubbc0\ub85c \ub354 \uc791\uc740 \uae30\uae30\ub098 \uc5e3\uc9c0 \ucef4\ud4e8\ud305 \uc2dc\ub098\ub9ac\uc624\uc5d0 \ubc30\ud3ec\ud558\ub294 \ub370 \uc801\ud569\ud569\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \uc628\ub514\ubc14\uc774\uc2a4 \ucc57\ubd07 \ubc0f \uac1c\uc778\ud654\ub41c \ubaa8\ubc14\uc77c \uc5b4\uc2dc\uc2a4\ud134\ud2b8\uc640 \uac19\uc740 \uc2e4\uc81c \uc560\ud50c\ub9ac\ucf00\uc774\uc158\uc5d0 \ub300\ud55c \uae30\ud68c\uac00 \uc5f4\ub9bd\ub2c8\ub2e4.<\/span><\/li>\n<li id=\"ad86\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc811\uadfc\uc131: \ub9ac\uc18c\uc2a4 \uc694\uad6c \uc0ac\ud56d\uc774 \ub0ae\uae30 \ub54c\ubb38\uc5d0 SLM\uc740 \ub354 \uad11\ubc94\uc704\ud55c \uac1c\ubc1c\uc790\uc640 \uc870\uc9c1\uc5d0\uc11c \ub354 \uc27d\uac8c \uc811\uadfc\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 AI\uac00 \ubbfc\uc8fc\ud654\ub418\uc5b4 \uc18c\uaddc\ubaa8 \ud300\uacfc \uac1c\ubcc4 \uc5f0\uad6c\uc790\uac00 \uc0c1\ub2f9\ud55c \uc778\ud504\ub77c \ud22c\uc790 \uc5c6\uc774 \uc5b8\uc5b4 \ubaa8\ub378\uc758 \ud798\uc744 \ud0d0\uad6c\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/span><\/li>\n<li id=\"3f9f\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc0ac\uc6a9\uc790 \uc815\uc758: SLM\uc740 \ud2b9\uc815 \ub3c4\uba54\uc778 \ubc0f \uc791\uc5c5\uc5d0 \ub300\ud574 \ubbf8\uc138 \uc870\uc815\ud558\uae30\uac00 \ub354 \uc27d\uc2b5\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \ud2c8\uc0c8 \uc2dc\uc7a5 \uc560\ud50c\ub9ac\ucf00\uc774\uc158\uc5d0 \ub9de\uac8c \uc870\uc815\ub41c \uc804\ubb38 \ubaa8\ub378\uc744 \ub9cc\ub4e4 \uc218 \uc788\uc5b4 \ub354 \ub192\uc740 \uc131\ub2a5\uacfc \uc815\ud655\uc131\uc744 \uc5bb\uc744 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/span><\/li>\n<\/ul>\n<h1 id=\"229c\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc740 \uc5b4\ub5bb\uac8c \uc791\ub3d9\ud558\ub098\uc694?<\/span><\/h1>\n<p id=\"5c44\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">LLM\uacfc \ub9c8\ucc2c\uac00\uc9c0\ub85c SLM\uc740 \ubc29\ub300\ud55c \ud14d\uc2a4\ud2b8 \ubc0f \ucf54\ub4dc \ub370\uc774\ud130 \uc138\ud2b8\uc5d0\uc11c \ud559\uc2b5\ub429\ub2c8\ub2e4. \uadf8\ub7ec\ub098 \ub354 \uc791\uc740 \ud06c\uae30\uc640 \ud6a8\uc728\uc131\uc744 \ub2ec\uc131\ud558\uae30 \uc704\ud574 \uc5ec\ub7ec \uac00\uc9c0 \uae30\uc220\uc774 \uc0ac\uc6a9\ub429\ub2c8\ub2e4.<\/span><\/p>\n<ul>\n<li id=\"e90e\" class=\"mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc9c0\uc2dd \uc99d\ub958: \uc774\ub294 \uc0ac\uc804 \ud6c8\ub828\ub41c LLM\uc5d0\uc11c \ub354 \uc791\uc740 \ubaa8\ub378\ub85c \uc9c0\uc2dd\uc744 \uc804\ub2ec\ud558\uc5ec \uc804\uccb4\uc801\uc778 \ubcf5\uc7a1\uc131\uc744 \uc81c\uac70\ud55c \ucc44 \ud575\uc2ec \uc5ed\ub7c9\uc744 \ud3ec\ucc29\ud558\ub294 \uac83\uc744 \ud3ec\ud568\ud569\ub2c8\ub2e4.<\/span><\/li>\n<li id=\"b0f9\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uac00\uc9c0\uce58\uae30 \ubc0f \uc591\uc790\ud654: \uc774\ub7ec\ud55c \uae30\uc220\uc740 \ubaa8\ub378\uc758 \ubd88\ud544\uc694\ud55c \ubd80\ubd84\uc744 \uc81c\uac70\ud558\uace0 \uac00\uc911\uce58\uc758 \uc815\ud655\ub3c4\ub97c \uac01\uac01 \ub0ae\ucd94\uc5b4 \ubaa8\ub378\uc758 \ud06c\uae30\uc640 \ub9ac\uc18c\uc2a4 \uc694\uad6c \uc0ac\ud56d\uc744 \ub354\uc6b1 \uc904\uc785\ub2c8\ub2e4.<\/span><\/li>\n<li id=\"0c7f\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb od oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\ud6a8\uc728\uc801\uc778 \uc544\ud0a4\ud14d\ucc98: \uc5f0\uad6c\uc790\ub4e4\uc740 SLM\uc744 \uc704\ud574 \ud2b9\ubcc4\ud788 \uc124\uacc4\ub41c \uc0c8\ub85c\uc6b4 \uc544\ud0a4\ud14d\ucc98\ub97c \uc9c0\uc18d\uc801\uc73c\ub85c \uac1c\ubc1c\ud558\uace0 \uc788\uc73c\uba70, \uc131\ub2a5\uacfc \ud6a8\uc728\uc131\uc744 \ubaa8\ub450 \ucd5c\uc801\ud654\ud558\ub294 \ub370 \uc911\uc810\uc744 \ub450\uace0 \uc788\uc2b5\ub2c8\ub2e4.<\/span><\/li>\n<\/ul>\n<h1 id=\"8ebc\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc774\uc810 \ubc0f \uc81c\ud55c \uc0ac\ud56d<\/span><\/h1>\n<p id=\"cf6e\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM)\uc740 \ube44\uad50\uc801 \uc801\uc740 \ub370\uc774\ud130 \uc138\ud2b8\ub85c \ud6c8\ub828\ud560 \uc218 \uc788\ub2e4\ub294 \uc7a5\uc810\uc774 \uc788\uc2b5\ub2c8\ub2e4. \uac04\uc18c\ud654\ub41c \uc544\ud0a4\ud14d\ucc98\ub294 \ud574\uc11d \uac00\ub2a5\uc131\uc744 \ub192\uc774\uace0, \ucef4\ud329\ud2b8\ud55c \ud06c\uae30\ub294 \ubaa8\ubc14\uc77c \uae30\uae30\uc5d0 \ubc30\ud3ec\ud558\ub294 \ub370 \uc6a9\uc774\ud569\ub2c8\ub2e4.<\/span><\/p>\n<p id=\"0699\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">SLM\uc758 \uc8fc\uc694 \uc774\uc810\uc740 \ub370\uc774\ud130\ub97c \ub85c\uceec\uc5d0\uc11c \ucc98\ub9ac\ud560 \uc218 \uc788\ub294 \uae30\ub2a5\uc73c\ub85c, \ud2b9\ud788 \uc0ac\ubb3c \uc778\ud130\ub137(IoT) \uc5d0\uc9c0 \uc7a5\uce58\uc640 \uc5c4\uaca9\ud55c \uac1c\uc778 \uc815\ubcf4 \ubcf4\ud638 \ubc0f \ubcf4\uc548 \uaddc\uc815\uc744 \uc900\uc218\ud574\uc57c \ud558\ub294 \uae30\uc5c5\uc5d0 \ub9e4\uc6b0 \uc720\uc6a9\ud569\ub2c8\ub2e4.<\/span><\/p>\n<p id=\"da95\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uadf8\ub7ec\ub098 \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc744 \ubc30\ud3ec\ud558\ub294 \ub370\ub294 \ud2b8\ub808\uc774\ub4dc\uc624\ud504\uac00 \uc218\ubc18\ub429\ub2c8\ub2e4. \uc18c\uaddc\ubaa8 \ub370\uc774\ud130 \uc138\ud2b8\uc5d0 \ub300\ud55c \ud6c8\ub828\uc73c\ub85c \uc778\ud574 SLM\uc740 \ub300\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(LLM) \ub300\uc751 \ubaa8\ub378\uc5d0 \ube44\ud574 \ub354 \uc81c\ud55c\ub41c \uc9c0\uc2dd \uae30\ubc18\uc744 \uac00\uc9c0\uace0 \uc788\uc2b5\ub2c8\ub2e4. \ub610\ud55c \uc5b8\uc5b4\uc640 \ub9e5\ub77d\uc5d0 \ub300\ud55c \uc774\ud574\uac00 \ub354 \uc81c\ud55c\ub418\ub294 \uacbd\ud5a5\uc774 \uc788\uc5b4 \ub300\uaddc\ubaa8 \ubaa8\ub378\uc5d0 \ube44\ud574 \ub35c \uc815\ud655\ud558\uace0 \ubbf8\ubb18\ud55c \uc751\ub2f5\uc774 \ub098\uc62c \uac00\ub2a5\uc131\uc774 \uc788\uc2b5\ub2c8\ub2e4.<\/span><\/p>\n<figure class=\"lv lw lx ly lz ma ls lt paragraph-image\">\n<div class=\"mb mc ed md bh me\" tabindex=\"0\" role=\"button\">\n<div class=\"ls lt ol\"><span style=\"color: #000000;\"><picture><source srcset=\"https:\/\/miro.medium.com\/v2\/resize:fit:640\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 640w, https:\/\/miro.medium.com\/v2\/resize:fit:720\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 720w, https:\/\/miro.medium.com\/v2\/resize:fit:750\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 750w, https:\/\/miro.medium.com\/v2\/resize:fit:786\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 786w, https:\/\/miro.medium.com\/v2\/resize:fit:828\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 828w, https:\/\/miro.medium.com\/v2\/resize:fit:1100\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1100w, https:\/\/miro.medium.com\/v2\/resize:fit:1400\/format:webp\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1400w\" type=\"image\/webp\" sizes=\"(min-resolution: 4dppx) and (max-width: 700px) 50vw, (-webkit-min-device-pixel-ratio: 4) and (max-width: 700px) 50vw, (min-resolution: 3dppx) and (max-width: 700px) 67vw, (-webkit-min-device-pixel-ratio: 3) and (max-width: 700px) 65vw, (min-resolution: 2.5dppx) and (max-width: 700px) 80vw, (-webkit-min-device-pixel-ratio: 2.5) and (max-width: 700px) 80vw, (min-resolution: 2dppx) and (max-width: 700px) 100vw, (-webkit-min-device-pixel-ratio: 2) and (max-width: 700px) 100vw, 700px\"><source srcset=\"https:\/\/miro.medium.com\/v2\/resize:fit:640\/1*YyfhMNzNOcFzpF-YoRWy9w.png 640w, https:\/\/miro.medium.com\/v2\/resize:fit:720\/1*YyfhMNzNOcFzpF-YoRWy9w.png 720w, https:\/\/miro.medium.com\/v2\/resize:fit:750\/1*YyfhMNzNOcFzpF-YoRWy9w.png 750w, https:\/\/miro.medium.com\/v2\/resize:fit:786\/1*YyfhMNzNOcFzpF-YoRWy9w.png 786w, https:\/\/miro.medium.com\/v2\/resize:fit:828\/1*YyfhMNzNOcFzpF-YoRWy9w.png 828w, https:\/\/miro.medium.com\/v2\/resize:fit:1100\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1100w, https:\/\/miro.medium.com\/v2\/resize:fit:1400\/1*YyfhMNzNOcFzpF-YoRWy9w.png 1400w\" sizes=\"(min-resolution: 4dppx) and (max-width: 700px) 50vw, (-webkit-min-device-pixel-ratio: 4) and (max-width: 700px) 50vw, (min-resolution: 3dppx) and (max-width: 700px) 67vw, (-webkit-min-device-pixel-ratio: 3) and (max-width: 700px) 65vw, (min-resolution: 2.5dppx) and (max-width: 700px) 80vw, (-webkit-min-device-pixel-ratio: 2.5) and (max-width: 700px) 80vw, (min-resolution: 2dppx) and (max-width: 700px) 100vw, (-webkit-min-device-pixel-ratio: 2) and (max-width: 700px) 100vw, 700px\" data-testid=\"og\"><img loading=\"lazy\" decoding=\"async\" class=\"bh la mf c\" role=\"presentation\" src=\"https:\/\/stechstar.com\/user\/zbxe\/files\/attach\/images\/92879\/020\/104\/de29769bbc3ce7b4fc50af0e1e6eddcd.png\" alt=\"\" width=\"700\" height=\"296\" data-autoattach=\"success\" data-cke-saved-src=\".\/files\/attach\/images\/92879\/020\/104\/de29769bbc3ce7b4fc50af0e1e6eddcd.png\"><\/picture><\/span><\/div>\n<\/div><figcaption class=\"om on oo ls lt op oq bf b bg z dx\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">SLM\uacfc LLM\uc758 \ube44\uad50<\/span><\/figcaption><\/figure>\n<h1 id=\"47bf\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM)\uc758 \uba87 \uac00\uc9c0 \uc608<\/span><\/h1>\n<ol>\n<li id=\"2117\" class=\"mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">DistilBERT<\/span>&nbsp;: DistilBERT\ub294 \uc790\uc5f0\uc5b4 \ucc98\ub9ac(NLP)\uc758 \uc120\uad6c\uc801 \ubaa8\ub378\uc778 BERT\uc758 \ubcf4\ub2e4 \ucef4\ud329\ud2b8\ud558\uace0 \ubbfc\ucca9\ud558\uba70 \uac00\ubcbc\uc6b4 \ubc18\ubcf5\uc744 \ub098\ud0c0\ub0c5\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/distilbert\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/distilbert\">https:\/\/huggingface.co\/docs\/transformers\/model_doc\/distilbert<\/a><\/span><\/li>\n<li id=\"ed72\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">Orca 2<\/span>&nbsp;: Microsoft\uc5d0\uc11c \uac1c\ubc1c\ud55c Orca 2\ub294 \uace0\ud488\uc9c8 \ud569\uc131 \ub370\uc774\ud130\ub97c \uc0ac\uc6a9\ud558\uc5ec Meta\uc758 Llama 2\ub97c \ubbf8\uc138 \uc870\uc815\ud55c \uacb0\uacfc\uc785\ub2c8\ub2e4. \uc774 \ud601\uc2e0\uc801\uc778 \uc811\uadfc \ubc29\uc2dd\uc744 \ud1b5\ud574 Microsoft\ub294 \ud2b9\ud788 \uc81c\ub85c\uc0f7 \ucd94\ub860 \uc791\uc5c5\uc5d0\uc11c \ub354 \ud070 \ubaa8\ub378\uacfc \uacbd\uc7c1\ud558\uac70\ub098 \ub2a5\uac00\ud558\ub294 \uc131\ub2a5 \uc218\uc900\uc744 \ub2ec\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/microsoft\/Orca-2-13b\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/microsoft\/Orca-2-13b\">https:\/\/huggingface.co\/microsoft\/Orca-2-13b<\/a><\/span><\/li>\n<li id=\"e520\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">Phi 2<\/span>&nbsp;: Microsoft\uc758 Phi 2\ub294 \ud074\ub77c\uc6b0\ub4dc\uc640 \uc5e3\uc9c0 \ubc30\ud3ec \ubaa8\ub450\uc5d0\uc11c \ud6a8\uc728\uc131\uacfc \uc801\uc751\uc131\uc744 \uc704\ud574 \uc124\uacc4\ub41c \ud2b8\ub79c\uc2a4\ud3ec\uba38 \uae30\ubc18 Small Language Model(SLM)\uc785\ub2c8\ub2e4. Microsoft\uc5d0 \ub530\ub974\uba74 Phi 2\ub294 \uc218\ud559\uc801 \ucd94\ub860, \uc0c1\uc2dd, \uc5b8\uc5b4 \uc774\ud574 \ubc0f \ub17c\ub9ac\uc801 \ucd94\ub860\uacfc \uac19\uc740 \ub3c4\uba54\uc778\uc5d0\uc11c \ucd5c\ucca8\ub2e8 \uc131\ub2a5\uc744 \ubcf4\uc5ec\uc90d\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/main\/model_doc\/phi\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/main\/model_doc\/phi\">https:\/\/huggingface.co\/docs\/transformers\/main\/model_doc\/phi<\/a><\/span><\/li>\n<li id=\"bbae\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">BERT Mini, Small, Medium, Tiny<\/span>&nbsp;: Google\uc758 BERT \ubaa8\ub378\uc740 440\ub9cc \uac1c\uc758 \ub9e4\uac1c\ubcc0\uc218\uac00 \uc788\ub294 Mini\ubd80\ud130 4,100\ub9cc \uac1c\uc758 \ub9e4\uac1c\ubcc0\uc218\uac00 \uc788\ub294 Medium\uae4c\uc9c0 \ub2e4\uc591\ud55c \ub9ac\uc18c\uc2a4 \uc81c\uc57d\uc744 \uc218\uc6a9\ud560 \uc218 \uc788\ub294 \ucd95\uc18c \ubc84\uc804\uc73c\ub85c \uc81c\uacf5\ub429\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/prajjwal1\/bert-mini\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/prajjwal1\/bert-mini\">https:\/\/huggingface.co\/prajjwal1\/bert-mini<\/a><\/span><\/li>\n<li id=\"98e3\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">GPT-Neo \ubc0f GPT-J<\/span>&nbsp;: GPT-Neo \ubc0f GPT-J\ub294 OpenAI\uc758 GPT \ubaa8\ub378\uc744 \ucd95\uc18c\ud55c \ubc84\uc804\uc774\uba70, \ub354 \uc81c\ud55c\ub41c \uacc4\uc0b0 \ub9ac\uc18c\uc2a4\uac00 \uc788\ub294 \uc560\ud50c\ub9ac\ucf00\uc774\uc158 \uc2dc\ub098\ub9ac\uc624\uc5d0\uc11c \ub2e4\uc591\uc131\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/gpt_neo\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/gpt_neo\">https:\/\/huggingface.co\/docs\/transformers\/model_doc\/gpt_neo<\/a><\/span><\/li>\n<li id=\"89e2\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">MobileBERT<\/span>&nbsp;: \ubaa8\ubc14\uc77c \uae30\uae30\uc5d0 \ub9de\ucdb0 \uc81c\uc791\ub41c MobileBERT\ub294 \ubaa8\ubc14\uc77c \ucef4\ud4e8\ud305\uc758 \uc81c\uc57d \ub0b4\uc5d0\uc11c \uc131\ub2a5\uc744 \ucd5c\uc801\ud654\ud558\ub3c4\ub85d \ud2b9\ubcc4\ud788 \uc124\uacc4\ub418\uc5c8\uc2b5\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/mobilebert\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/docs\/transformers\/model_doc\/mobilebert\">https:\/\/huggingface.co\/docs\/transformers\/model_doc\/mobilebert<\/a><\/span><\/li>\n<li id=\"d89c\" class=\"mg mh fq mi b go og mk ml gr oh mn mo mp oi mr ms mt oj mv mw mx ok mz na nb or oe of bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\"><span class=\"mi fr\">T5-Small<\/span>&nbsp;: Google\uc758 Text-to-Text Transfer Transformer(T5) \ubaa8\ub378 \uc2dc\ub9ac\uc988\uc758 \uc77c\ubd80\uc778 T5-Small\uc740 \uc131\ub2a5\uacfc \ub9ac\uc18c\uc2a4 \ud65c\uc6a9 \uac04\uc758 \uade0\ud615\uc744 \ub9de\ucdb0 \ud6a8\uc728\uc801\uc778 \ud14d\uc2a4\ud2b8 \ucc98\ub9ac \uae30\ub2a5\uc744 \uc81c\uacf5\ud558\ub294 \uac83\uc744 \ubaa9\ud45c\ub85c \ud569\ub2c8\ub2e4. \u2014&nbsp;<a class=\"af os\" style=\"color: #000000;\" href=\"https:\/\/huggingface.co\/t5-small\" target=\"_blank\" rel=\"noopener ugc nofollow\" data-cke-saved-href=\"https:\/\/huggingface.co\/t5-small\">https:\/\/huggingface.co\/t5-small<\/a><\/span><\/li>\n<\/ol>\n<h1 id=\"cc53\" class=\"nc nd fq bf ne nf ng gq nh ni nj gt nk nl nm nn no np nq nr ns nt nu nv nw nx bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc758 \ubbf8\ub798<\/span><\/h1>\n<p id=\"0ff5\" class=\"pw-post-body-paragraph mg mh fq mi b go ny mk ml gr nz mn mo mp oa mr ms mt ob mv mw mx oc mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uc5f0\uad6c \uac1c\ubc1c\uc774 \uc9c4\ud589\ub428\uc5d0 \ub530\ub77c SLM\uc774 \ub354\uc6b1 \uac15\ub825\ud558\uace0 \ub2e4\uc7ac\ub2e4\ub2a5\ud574\uc9c8 \uac83\uc73c\ub85c \uc608\uc0c1\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \ud6c8\ub828 \uae30\uc220, \ud558\ub4dc\uc6e8\uc5b4 \ubc1c\uc804, \ud6a8\uc728\uc801\uc778 \uc544\ud0a4\ud14d\ucc98\uc758 \uac1c\uc120\uc73c\ub85c SLM\uacfc LLM \uac04\uc758 \uaca9\ucc28\ub294 \uacc4\uc18d \uc881\uc544\uc9c8 \uac83\uc785\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \uc0c8\ub86d\uace0 \ud765\ubbf8\ub85c\uc6b4 \uc751\uc6a9 \ubd84\uc57c\ub85c\uc758 \ubb38\uc774 \uc5f4\ub9ac\uace0 AI\uc640 \uadf8\uac83\uc774 \uc6b0\ub9ac \uc0b6\uc5d0 \uc601\ud5a5\uc744 \ubbf8\uce60 \uc7a0\uc7ac\ub825\uc774 \ub354\uc6b1 \ubbfc\uc8fc\ud654\ub420 \uac83\uc785\ub2c8\ub2e4.<\/span><\/p>\n<p id=\"f9c7\" class=\"pw-post-body-paragraph mg mh fq mi b go mj mk ml gr mm mn mo mp mq mr ms mt mu mv mw mx my mz na nb fj bk\" data-selectable-paragraph=\"\"><span style=\"color: #000000;\">\uacb0\ub860\uc801\uc73c\ub85c, \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc740 AI\uc758 \ud48d\uacbd\uc5d0\uc11c \uc0c1\ub2f9\ud55c \ubcc0\ud654\ub97c \ub098\ud0c0\ub0c5\ub2c8\ub2e4. \ud6a8\uc728\uc131, \uc811\uadfc\uc131 \ubc0f \uc0ac\uc6a9\uc790 \uc815\uc758 \uae30\ub2a5\uc740 \ub2e4\uc591\ud55c \ub3c4\uba54\uc778\uc758 \uac1c\ubc1c\uc790\uc640 \uc5f0\uad6c\uc790\uc5d0\uac8c \uadc0\uc911\ud55c \ub3c4\uad6c\uac00 \ub429\ub2c8\ub2e4. SLM\uc774 \uacc4\uc18d \uc9c4\ud654\ud568\uc5d0 \ub530\ub77c \uac1c\uc778\uacfc \uc870\uc9c1 \ubaa8\ub450\uc5d0\uac8c \ud798\uc744 \uc2e4\uc5b4 \uc904 \uc5c4\uccad\ub09c \uc57d\uc18d\uc744 \ub2f4\uace0 \uc788\uc73c\uba70, AI\uac00 \uac15\ub825\ud560 \ubfd0\ub9cc \uc544\ub2c8\ub77c \uc811\uadfc\uc131\uc774 \ub6f0\uc5b4\ub098\uace0 \ub2e4\uc591\ud55c \uc694\uad6c\uc5d0 \ub9de\uac8c \uc870\uc815\ub418\ub294 \ubbf8\ub798\ub97c \ud615\uc131\ud569\ub2c8\ub2e4.<\/span><\/p>\n<\/div>\n<div class=\"l ib\">&nbsp;<\/div>\n<div class=\"l ib\">&nbsp;<\/div>\n<div class=\"pvc_clear\"><\/div>\n<p id=\"pvc_stats_10943\" class=\"pvc_stats all  \" data-element-id=\"10943\" style=\"\"><i class=\"pvc-stats-icon medium\" aria-hidden=\"true\"><svg aria-hidden=\"true\" focusable=\"false\" data-prefix=\"far\" data-icon=\"chart-bar\" role=\"img\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 512 512\" class=\"svg-inline--fa fa-chart-bar fa-w-16 fa-2x\"><path fill=\"currentColor\" d=\"M396.8 352h22.4c6.4 0 12.8-6.4 12.8-12.8V108.8c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v230.4c0 6.4 6.4 12.8 12.8 12.8zm-192 0h22.4c6.4 0 12.8-6.4 12.8-12.8V140.8c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v198.4c0 6.4 6.4 12.8 12.8 12.8zm96 0h22.4c6.4 0 12.8-6.4 12.8-12.8V204.8c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v134.4c0 6.4 6.4 12.8 12.8 12.8zM496 400H48V80c0-8.84-7.16-16-16-16H16C7.16 64 0 71.16 0 80v336c0 17.67 14.33 32 32 32h464c8.84 0 16-7.16 16-16v-16c0-8.84-7.16-16-16-16zm-387.2-48h22.4c6.4 0 12.8-6.4 12.8-12.8v-70.4c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v70.4c0 6.4 6.4 12.8 12.8 12.8z\" class=\"\"><\/path><\/svg><\/i> <img loading=\"lazy\" decoding=\"async\" width=\"16\" height=\"16\" alt=\"Loading\" src=\"https:\/\/www.stechstar.com\/user\/wordpress\/wp-content\/plugins\/page-views-count\/ajax-loader-2x.gif\" border=0 \/><\/p>\n<div class=\"pvc_clear\"><\/div>\n","protected":false},"excerpt":{"rendered":"<p>[\uc778\uacf5\uc9c0\ub2a5 \uae30\uc220] \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378(SLM) \uc18c\uaddc\ubaa8 \uc5b8\uc5b4 \ubaa8\ub378\uc758 \ubd80\uc0c1: AI\ub97c \uc704\ud55c \ud6a8\uc728\uc131\uacfc \ub9de\ucda4\ud654 : Small Language Models (SLMs) Small Language Models (SLMs) The Rise of Small Language Models: Efficiency and Customization for AI Nagesh Mashette Large language models (LLMs) have captured headlines and imaginations with their impressive capabilities in natural language processing. However, their massive [&hellip;]<\/p>\n<div class=\"pvc_clear\"><\/div>\n<p id=\"pvc_stats_10943\" class=\"pvc_stats all  \" data-element-id=\"10943\" style=\"\"><i class=\"pvc-stats-icon medium\" aria-hidden=\"true\"><svg aria-hidden=\"true\" focusable=\"false\" data-prefix=\"far\" data-icon=\"chart-bar\" role=\"img\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 512 512\" class=\"svg-inline--fa fa-chart-bar fa-w-16 fa-2x\"><path fill=\"currentColor\" d=\"M396.8 352h22.4c6.4 0 12.8-6.4 12.8-12.8V108.8c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v230.4c0 6.4 6.4 12.8 12.8 12.8zm-192 0h22.4c6.4 0 12.8-6.4 12.8-12.8V140.8c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v198.4c0 6.4 6.4 12.8 12.8 12.8zm96 0h22.4c6.4 0 12.8-6.4 12.8-12.8V204.8c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v134.4c0 6.4 6.4 12.8 12.8 12.8zM496 400H48V80c0-8.84-7.16-16-16-16H16C7.16 64 0 71.16 0 80v336c0 17.67 14.33 32 32 32h464c8.84 0 16-7.16 16-16v-16c0-8.84-7.16-16-16-16zm-387.2-48h22.4c6.4 0 12.8-6.4 12.8-12.8v-70.4c0-6.4-6.4-12.8-12.8-12.8h-22.4c-6.4 0-12.8 6.4-12.8 12.8v70.4c0 6.4 6.4 12.8 12.8 12.8z\" class=\"\"><\/path><\/svg><\/i> <img loading=\"lazy\" decoding=\"async\" width=\"16\" height=\"16\" alt=\"Loading\" src=\"https:\/\/www.stechstar.com\/user\/wordpress\/wp-content\/plugins\/page-views-count\/ajax-loader-2x.gif\" border=0 \/><\/p>\n<div class=\"pvc_clear\"><\/div>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0,"footnotes":""},"categories":[15,25,21,16,20],"tags":[],"class_list":["post-10943","post","type-post","status-publish","format-standard","hentry","category-stechstar-com-","category-25","category-21","category-16","category-20"],"a3_pvc":{"activated":true,"total_views":29,"today_views":0},"_links":{"self":[{"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/posts\/10943","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/comments?post=10943"}],"version-history":[{"count":4,"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/posts\/10943\/revisions"}],"predecessor-version":[{"id":11739,"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/posts\/10943\/revisions\/11739"}],"wp:attachment":[{"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/media?parent=10943"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/categories?post=10943"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.stechstar.com\/user\/wordpress\/wp-json\/wp\/v2\/tags?post=10943"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}