{"id":72,"date":"2024-06-02T10:47:01","date_gmt":"2024-06-02T10:47:01","guid":{"rendered":"https:\/\/www.asm15.fr\/?p=72"},"modified":"2024-06-02T10:47:01","modified_gmt":"2024-06-02T10:47:01","slug":"pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive","status":"publish","type":"post","link":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/","title":{"rendered":"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive"},"content":{"rendered":"<p><img decoding=\"async\" src=\"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&#038;se=2024-05-25T10%3A45%3A25Z&#038;sp=r&#038;sv=2021-08-06&#038;sr=b&#038;rscd=inline&#038;rsct=image\/png&#038;skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&#038;sktid=a48cca56-e6da-484e-a814-9c849652bcb3&#038;skt=2024-05-24T19%3A17%3A34Z&#038;ske=2024-05-25T19%3A17%3A34Z&#038;sks=b&#038;skv=2021-08-06&#038;sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D\"><\/p>\n<p>Les mod\u00e8les de langage tels que GPT-4 excellent dans la compr\u00e9hension linguistique, mais rencontrent des probl\u00e8mes de m\u00e9moire GPU \u00e9lev\u00e9e pendant l&rsquo;inf\u00e9rence, limitant leur \u00e9volutivit\u00e9. Les m\u00e9thodes existantes r\u00e9duisent la m\u00e9moire en compressant le cache KV, mais n\u00e9gligent les d\u00e9pendances entre couches et les demandes de pr\u00e9-calcul. Par exemple, un mod\u00e8le de 7 milliards de param\u00e8tres utilise 14 Go pour les param\u00e8tres mais 72 Go pour le cache KV. Des chercheurs de l&rsquo;Universit\u00e9 de Shanghai Jiao Tong et d&rsquo;autres ont d\u00e9velopp\u00e9 PyramidInfer, qui compresse le cache KV en ne conservant que les cl\u00e9s et valeurs contextuelles cruciales. Des tests montrent que PyramidInfer am\u00e9liore le d\u00e9bit de 2,2 fois et r\u00e9duit la m\u00e9moire du cache de plus de 54%.<\/p>\n<p>Pour r\u00e9pondre \u00e0 la demande croissante des chatbots, des strat\u00e9gies efficaces sont essentielles. PyramidInfer se d\u00e9marque en consid\u00e9rant la compression sp\u00e9cifique aux couches. En v\u00e9rifiant les hypoth\u00e8ses ICR et RAC, PyramidInfer r\u00e9v\u00e8le une haute redondance dans les couches profondes, permettant une r\u00e9duction significative du cache KV sans affecter la qualit\u00e9 de sortie. Des \u00e9valuations sur divers mod\u00e8les confirment que PyramidInfer r\u00e9duit l&rsquo;utilisation de la m\u00e9moire GPU tout en maintenant la qualit\u00e9 de g\u00e9n\u00e9ration, faisant de cette approche une solution id\u00e9ale pour les environnements \u00e0 ressources limit\u00e9es.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Les mod\u00e8les de langage tels que GPT-4 excellent dans la compr\u00e9hension linguistique, mais rencontrent des probl\u00e8mes de m\u00e9moire GPU \u00e9lev\u00e9e pendant l&rsquo;inf\u00e9rence, limitant leur \u00e9volutivit\u00e9. Les m\u00e9thodes existantes r\u00e9duisent la m\u00e9moire en compressant le cache KV, mais n\u00e9gligent les d\u00e9pendances entre couches et les demandes de pr\u00e9-calcul. Par exemple, un mod\u00e8le de 7 milliards de [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-72","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v25.5 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive - IA &amp; TECHNOLOGIES<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive - IA &amp; TECHNOLOGIES\" \/>\n<meta property=\"og:description\" content=\"Les mod\u00e8les de langage tels que GPT-4 excellent dans la compr\u00e9hension linguistique, mais rencontrent des probl\u00e8mes de m\u00e9moire GPU \u00e9lev\u00e9e pendant l&rsquo;inf\u00e9rence, limitant leur \u00e9volutivit\u00e9. Les m\u00e9thodes existantes r\u00e9duisent la m\u00e9moire en compressant le cache KV, mais n\u00e9gligent les d\u00e9pendances entre couches et les demandes de pr\u00e9-calcul. Par exemple, un mod\u00e8le de 7 milliards de [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/\" \/>\n<meta property=\"og:site_name\" content=\"IA &amp; TECHNOLOGIES\" \/>\n<meta property=\"article:published_time\" content=\"2024-06-02T10:47:01+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D\" \/>\n<meta name=\"author\" content=\"m\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"m\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"1 minute\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/\",\"url\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/\",\"name\":\"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive - IA &amp; TECHNOLOGIES\",\"isPartOf\":{\"@id\":\"https:\/\/www.asm15.fr\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D\",\"datePublished\":\"2024-06-02T10:47:01+00:00\",\"author\":{\"@id\":\"https:\/\/www.asm15.fr\/#\/schema\/person\/9e2a0039267955de05a7ce4282a905e6\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#primaryimage\",\"url\":\"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D\",\"contentUrl\":\"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Accueil\",\"item\":\"https:\/\/www.asm15.fr\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.asm15.fr\/#website\",\"url\":\"https:\/\/www.asm15.fr\/\",\"name\":\"IA & TECHNOLOGIES\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.asm15.fr\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.asm15.fr\/#\/schema\/person\/9e2a0039267955de05a7ce4282a905e6\",\"name\":\"m\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/www.asm15.fr\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/6fe5cecb08501f51d81efa38bdd65daa775b9224f6dc9617c8615be766310093?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/6fe5cecb08501f51d81efa38bdd65daa775b9224f6dc9617c8615be766310093?s=96&d=mm&r=g\",\"caption\":\"m\"},\"sameAs\":[\"https:\/\/www.asm15.fr\"],\"url\":\"https:\/\/www.asm15.fr\/index.php\/author\/m\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive - IA &amp; TECHNOLOGIES","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/","og_locale":"fr_FR","og_type":"article","og_title":"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive - IA &amp; TECHNOLOGIES","og_description":"Les mod\u00e8les de langage tels que GPT-4 excellent dans la compr\u00e9hension linguistique, mais rencontrent des probl\u00e8mes de m\u00e9moire GPU \u00e9lev\u00e9e pendant l&rsquo;inf\u00e9rence, limitant leur \u00e9volutivit\u00e9. Les m\u00e9thodes existantes r\u00e9duisent la m\u00e9moire en compressant le cache KV, mais n\u00e9gligent les d\u00e9pendances entre couches et les demandes de pr\u00e9-calcul. Par exemple, un mod\u00e8le de 7 milliards de [&hellip;]","og_url":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/","og_site_name":"IA &amp; TECHNOLOGIES","article_published_time":"2024-06-02T10:47:01+00:00","og_image":[{"url":"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D","type":"","width":"","height":""}],"author":"m","twitter_card":"summary_large_image","twitter_misc":{"\u00c9crit par":"m","Dur\u00e9e de lecture estim\u00e9e":"1 minute"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/","url":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/","name":"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive - IA &amp; TECHNOLOGIES","isPartOf":{"@id":"https:\/\/www.asm15.fr\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#primaryimage"},"image":{"@id":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#primaryimage"},"thumbnailUrl":"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D","datePublished":"2024-06-02T10:47:01+00:00","author":{"@id":"https:\/\/www.asm15.fr\/#\/schema\/person\/9e2a0039267955de05a7ce4282a905e6"},"breadcrumb":{"@id":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#primaryimage","url":"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D","contentUrl":"https:\/\/oaidalleapiprodscus.blob.core.windows.net\/private\/org-PXteLUy4B6mslwRon6h7ri9z\/user-0jfma6aPNvZyLXZUbw4CmvZG\/img-b1YihjUg62FQJEW4h94iXmiy.png?st=2024-05-25T08%3A45%3A25Z&se=2024-05-25T10%3A45%3A25Z&sp=r&sv=2021-08-06&sr=b&rscd=inline&rsct=image\/png&skoid=6aaadede-4fb3-4698-a8f6-684d7786b067&sktid=a48cca56-e6da-484e-a814-9c849652bcb3&skt=2024-05-24T19%3A17%3A34Z&ske=2024-05-25T19%3A17%3A34Z&sks=b&skv=2021-08-06&sig=eENwVOwgfNd0dANeQO2oDd2rIHhjPDQ3zncAJBeFolM%3D"},{"@type":"BreadcrumbList","@id":"https:\/\/www.asm15.fr\/index.php\/2024\/06\/02\/pyramidinfer-optimisation-efficace-de-la-compression-kv-cache-pour-une-inference-llm-evolutive\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Accueil","item":"https:\/\/www.asm15.fr\/"},{"@type":"ListItem","position":2,"name":"PyramidInfer : Optimisation Efficace de la Compression KV Cache pour une Inf\u00e9rence LLM \u00c9volutive"}]},{"@type":"WebSite","@id":"https:\/\/www.asm15.fr\/#website","url":"https:\/\/www.asm15.fr\/","name":"IA & TECHNOLOGIES","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.asm15.fr\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"},{"@type":"Person","@id":"https:\/\/www.asm15.fr\/#\/schema\/person\/9e2a0039267955de05a7ce4282a905e6","name":"m","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/www.asm15.fr\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/6fe5cecb08501f51d81efa38bdd65daa775b9224f6dc9617c8615be766310093?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/6fe5cecb08501f51d81efa38bdd65daa775b9224f6dc9617c8615be766310093?s=96&d=mm&r=g","caption":"m"},"sameAs":["https:\/\/www.asm15.fr"],"url":"https:\/\/www.asm15.fr\/index.php\/author\/m\/"}]}},"_links":{"self":[{"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/posts\/72","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/comments?post=72"}],"version-history":[{"count":1,"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/posts\/72\/revisions"}],"predecessor-version":[{"id":143,"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/posts\/72\/revisions\/143"}],"wp:attachment":[{"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/media?parent=72"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/categories?post=72"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.asm15.fr\/index.php\/wp-json\/wp\/v2\/tags?post=72"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}