{"id":2159,"date":"2026-04-01T13:01:43","date_gmt":"2026-04-01T11:01:43","guid":{"rendered":"https:\/\/askem.eu\/?p=2159"},"modified":"2026-04-01T13:01:49","modified_gmt":"2026-04-01T11:01:49","slug":"qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique","status":"publish","type":"post","link":"https:\/\/askem.eu\/en\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/","title":{"rendered":"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">Qdrant&nbsp;: base vectorielle open source pour le RAG et la recherche s\u00e9mantique<\/h2>\n\n\n\n<p>Les mod\u00e8les de langage ne retiennent pas vos donn\u00e9es. Pour que votre LLM r\u00e9ponde avec pr\u00e9cision sur un corpus interne&nbsp;: documents, notices, bases de connaissances, il faut lui fournir du contexte au moment de la requ\u00eate. C&rsquo;est le principe du RAG (<em>Retrieval-Augmented Generation<\/em>). Et au c\u0153ur de tout pipeline RAG, il y a une base vectorielle. Qdrant est aujourd&rsquo;hui l&rsquo;une des solutions open source les plus matures&nbsp;: performante, auto-h\u00e9bergeable, et dot\u00e9e d&rsquo;une API REST claire.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Pourquoi une base vectorielle<\/h2>\n\n\n\n<p>Un texte transform\u00e9 en vecteur (embedding) capture sa s\u00e9mantique sous forme d&rsquo;un tableau de nombres. Deux phrases proches s\u00e9mantiquement auront des vecteurs proches dans cet espace, m\u00eame si elles n&rsquo;ont aucun mot en commun. Une base vectorielle stocke ces vecteurs et permet de retrouver, pour une requ\u00eate donn\u00e9e, les <em>k<\/em> vecteurs les plus proches par similarit\u00e9 cosinus ou distance euclidienne. C&rsquo;est la brique de recherche au fondement des pipelines RAG modernes.<\/p>\n\n\n\n<p>Les bases relationnelles classiques ne sont pas con\u00e7ues pour cette op\u00e9ration&nbsp;: une recherche par similarit\u00e9 sur des millions de vecteurs de haute dimension requiert des index sp\u00e9cialis\u00e9s (HNSW, IVF). Qdrant impl\u00e9mente HNSW en natif, avec des performances adapt\u00e9es \u00e0 la production.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Pr\u00e9sentation de Qdrant<\/h2>\n\n\n\n<p><a href=\"https:\/\/qdrant.tech\" target=\"_blank\" rel=\"noreferrer noopener\">Qdrant<\/a> est une base vectorielle \u00e9crite en Rust, publi\u00e9e sous licence Apache 2.0. Elle expose une API REST et une API gRPC, un client Python officiel, et une interface web l\u00e9g\u00e8re pour inspecter collections et points. Elle prend en charge les vecteurs denses (embeddings standards), les vecteurs \u00e9pars (BM25-style) et le mode <em>sparse + dense<\/em> combin\u00e9, appel\u00e9 <em>hybrid search<\/em>.<\/p>\n\n\n\n<p>Qdrant stocke, avec chaque vecteur, un <em>payload<\/em> JSON arbitraire \u2014 titre, URL, horodatage, cat\u00e9gorie \u2014 filtrable lors des requ\u00eates. Cela permet de combiner filtre m\u00e9tadonn\u00e9es + similarit\u00e9 vectorielle dans une seule requ\u00eate, sans post-traitement applicatif.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">D\u00e9ploiement Docker<\/h2>\n\n\n\n<p>L&rsquo;installation la plus directe repose sur Docker&nbsp;:<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">docker run -d \\\n  --name qdrant \\\n  -p 6333:6333 \\\n  -p 6334:6334 \\\n  -v qdrant_storage:\/qdrant\/storage \\\n  qdrant\/qdrant\n<\/pre>\n\n\n\n<p>Le port 6333 expose l&rsquo;API REST, le port 6334 l&rsquo;API gRPC. L&rsquo;interface web est accessible \u00e0 <code>http:\/\/localhost:6333\/dashboard<\/code>. Pour une stack Docker Compose, ajoutez simplement le service qdrant avec un volume nomm\u00e9 et un <code>restart: unless-stopped<\/code>.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Variables de configuration utiles<\/h3>\n\n\n\n<p>Qdrant se configure via un fichier <code>config\/production.yaml<\/code> ou par variables d&rsquo;environnement. Les param\u00e8tres les plus utiles en production&nbsp;:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><code>QDRANT__SERVICE__API_KEY<\/code> \u2014 cl\u00e9 d&rsquo;API pour prot\u00e9ger l&rsquo;acc\u00e8s<\/li>\n\n\n\n<li><code>QDRANT__STORAGE__WAL__WAL_CAPACITY_MB<\/code> \u2014 capacit\u00e9 du write-ahead log<\/li>\n\n\n\n<li><code>QDRANT__STORAGE__ON_DISK_PAYLOAD<\/code> \u2014 stockage du payload sur disque (m\u00e9moire r\u00e9duite)<\/li>\n\n\n\n<li><code>QDRANT__CLUSTER__ENABLED<\/code> \u2014 mode cluster distribu\u00e9 (disponible en version enterprise)<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">Cr\u00e9er une collection et indexer des documents<\/h2>\n\n\n\n<p>Une <em>collection<\/em> regroupe des vecteurs de m\u00eame dimension. Voici un exemple complet en Python, en combinant Qdrant avec un mod\u00e8le d&#8217;embedding local via Ollama (voir <a href=\"https:\/\/askem.eu\/2026\/03\/29\/ollama-executer-des-llm-en-local\/\">l&rsquo;article sur Ollama<\/a>)&nbsp;:<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">from qdrant_client import QdrantClient\nfrom qdrant_client.models import Distance, VectorParams, PointStruct\nimport ollama\n\nclient = QdrantClient(url=\"http:\/\/localhost:6333\")\n\n# Cr\u00e9ation de la collection (1 536 dimensions pour nomic-embed-text)\nclient.recreate_collection(\n    collection_name=\"documents\",\n    vectors_config=VectorParams(size=768, distance=Distance.COSINE),\n)\n\n# Indexation de quelques textes\ntextes = [\n    {\"id\": 1, \"texte\": \"CKAN est un portail de donn\u00e9es open source.\", \"titre\": \"CKAN\"},\n    {\"id\": 2, \"texte\": \"Keycloak g\u00e8re l'authentification SSO.\", \"titre\": \"Keycloak\"},\n    {\"id\": 3, \"texte\": \"Qdrant stocke des vecteurs pour la recherche s\u00e9mantique.\", \"titre\": \"Qdrant\"},\n]\n\npoints = []\nfor doc in textes:\n    embedding = ollama.embeddings(model=\"nomic-embed-text\", prompt=doc[\"texte\"])[\"embedding\"]\n    points.append(PointStruct(\n        id=doc[\"id\"],\n        vector=embedding,\n        payload={\"titre\": doc[\"titre\"], \"texte\": doc[\"texte\"]},\n    ))\n\nclient.upsert(collection_name=\"documents\", points=points)\nprint(f\"{len(points)} documents index\u00e9s.\")\n<\/pre>\n\n\n\n<h2 class=\"wp-block-heading\">Requ\u00eate de recherche s\u00e9mantique<\/h2>\n\n\n\n<p>Pour retrouver les documents les plus pertinents face \u00e0 une question&nbsp;:<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\">question = \"Comment g\u00e9rer les identit\u00e9s dans ma stack&nbsp;?\"\nquery_vector = ollama.embeddings(model=\"nomic-embed-text\", prompt=question)[\"embedding\"]\n\nr\u00e9sultats = client.search(\n    collection_name=\"documents\",\n    query_vector=query_vector,\n    limit=3,\n    with_payload=True,\n)\n\nfor r in r\u00e9sultats:\n    print(f\"Score {r.score:.3f} \u2014 {r.payload['titre']}&nbsp;: {r.payload['texte']}\")\n<\/pre>\n\n\n\n<p>Qdrant retourne les documents tri\u00e9s par score de similarit\u00e9. Ces passages peuvent \u00eatre directement inject\u00e9s dans le prompt envoy\u00e9 \u00e0 votre LLM pour former un pipeline RAG complet.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Hybrid search&nbsp;: combiner s\u00e9mantique et lexical<\/h2>\n\n\n\n<p>La recherche purement vectorielle peut manquer des correspondances exactes sur des termes techniques (noms propres, acronymes, identifiants). Le mode <em>hybrid search<\/em> de Qdrant combine un vecteur dense (embedding) et un vecteur \u00e9pars (repr\u00e9sentation BM25-like via <code>sparse vectors<\/code>) dans la m\u00eame requ\u00eate, avec pond\u00e9ration configurable. Pour les corpus techniques&nbsp;: documentation, CKAN, tickets, ce mode am\u00e9liore significativement la pertinence.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Int\u00e9gration dans un pipeline RAG complet<\/h2>\n\n\n\n<p>L&rsquo;architecture typique avec les outils couverts sur ce site&nbsp;:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Sources&nbsp;:<\/strong> fichiers CSV\/JSON depuis CKAN, documents Nextcloud, exports PostgreSQL<\/li>\n\n\n\n<li><strong>Embedding&nbsp;:<\/strong> Ollama (<code>nomic-embed-text<\/code>, <code>mxbai-embed-large<\/code>) ou sentence-transformers en local<\/li>\n\n\n\n<li><strong>Stockage&nbsp;:<\/strong> Qdrant (base vectorielle) \u2014 les chunks de texte avec leurs m\u00e9tadonn\u00e9es<\/li>\n\n\n\n<li><strong>Retrieval&nbsp;:<\/strong> requ\u00eate s\u00e9mantique + filtre payload (date, source, cat\u00e9gorie)<\/li>\n\n\n\n<li><strong>G\u00e9n\u00e9ration&nbsp;:<\/strong> LLM local via Ollama (Llama 3, Mistral, Gemma) avec le contexte r\u00e9cup\u00e9r\u00e9<\/li>\n\n\n\n<li><strong>Orchestration&nbsp;:<\/strong> LangChain, LlamaIndex, ou n8n pour assembler le pipeline<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">Persistance, snapshots et sauvegarde<\/h2>\n\n\n\n<p>Qdrant stocke ses donn\u00e9es dans le volume mont\u00e9 (<code>\/qdrant\/storage<\/code>). Pour les sauvegardes, l&rsquo;API fournit un endpoint de snapshot&nbsp;:<\/p>\n\n\n\n<pre class=\"wp-block-preformatted\"># Cr\u00e9er un snapshot de la collection \"documents\"\ncurl -X POST http:\/\/localhost:6333\/collections\/documents\/snapshots\n\n# Lister les snapshots disponibles\ncurl http:\/\/localhost:6333\/collections\/documents\/snapshots\n<\/pre>\n\n\n\n<p>Les fichiers de snapshot peuvent \u00eatre envoy\u00e9s sur MinIO, S3 ou sauvegard\u00e9s avec BorgBackup (voir l&rsquo;article d\u00e9di\u00e9). La restauration se fait via le m\u00eame endpoint avec une requ\u00eate POST sur <code>\/collections\/{name}\/snapshots\/recover<\/code>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Dimensionnement et performance<\/h2>\n\n\n\n<p>Quelques r\u00e8gles pratiques pour calibrer votre d\u00e9ploiement Qdrant&nbsp;:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>1 million de vecteurs de 768 dimensions occupent environ 3 Go en RAM avec l&rsquo;index HNSW en m\u00e9moire<\/li>\n\n\n\n<li>Pour r\u00e9duire l&#8217;empreinte m\u00e9moire, activez <code>on_disk_payload: true<\/code> et le quantization (scalar ou product)<\/li>\n\n\n\n<li>Le param\u00e8tre <code>m<\/code> de l&rsquo;index HNSW contr\u00f4le le rapport qualit\u00e9\/m\u00e9moire (d\u00e9faut&nbsp;: 16)<\/li>\n\n\n\n<li>Pour les cas d&rsquo;usage RAG courants (&lt; 500 000 chunks), une instance single-node suffit largement<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">Alternatives open source<\/h2>\n\n\n\n<p>Qdrant n&rsquo;est pas la seule base vectorielle open source. <strong>Chroma<\/strong> est plus simple \u00e0 d\u00e9marrer (pens\u00e9 pour le prototypage Python). <strong>Weaviate<\/strong> inclut des modules d&#8217;embedding int\u00e9gr\u00e9s mais est plus lourd. <strong>Milvus<\/strong> est orient\u00e9 scalabilit\u00e9 extr\u00eame avec une architecture distribu\u00e9e. <strong>pgvector<\/strong> (extension PostgreSQL) convient si vous souhaitez \u00e9viter un service suppl\u00e9mentaire et que votre volume de vecteurs reste mod\u00e9r\u00e9. Pour une stack auto-h\u00e9berg\u00e9e en production, Qdrant offre le meilleur \u00e9quilibre entre performance, l\u00e9g\u00e8ret\u00e9 et maturit\u00e9 op\u00e9rationnelle.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Conclusion<\/h2>\n\n\n\n<p>Qdrant compl\u00e8te naturellement une stack AI open source&nbsp;: aux c\u00f4t\u00e9s d&rsquo;Ollama pour l&rsquo;inf\u00e9rence et les embeddings, de n8n pour l&rsquo;orchestration, et de PostgreSQL ou CKAN comme sources de donn\u00e9es, il forme le pivot de tout pipeline RAG auto-h\u00e9berg\u00e9. Son API claire, ses snapshots int\u00e9gr\u00e9s et sa faible empreinte op\u00e9rationnelle en font un choix solide d\u00e8s les premiers prototypes jusqu&rsquo;\u00e0 la mise en production.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Qdrant&nbsp;: base vectorielle open source pour le RAG et la recherche s\u00e9mantique Les mod\u00e8les de langage ne retiennent pas vos donn\u00e9es. Pour que votre LLM r\u00e9ponde avec pr\u00e9cision sur un corpus interne&nbsp;: documents, notices, bases de connaissances, il faut lui fournir du contexte au moment de la requ\u00eate. C&rsquo;est le principe du RAG (Retrieval-Augmented Generation). [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":2160,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ocean_post_layout":"","ocean_both_sidebars_style":"","ocean_both_sidebars_content_width":0,"ocean_both_sidebars_sidebars_width":0,"ocean_sidebar":"","ocean_second_sidebar":"","ocean_disable_margins":"enable","ocean_add_body_class":"","ocean_shortcode_before_top_bar":"","ocean_shortcode_after_top_bar":"","ocean_shortcode_before_header":"","ocean_shortcode_after_header":"","ocean_has_shortcode":"","ocean_shortcode_after_title":"","ocean_shortcode_before_footer_widgets":"","ocean_shortcode_after_footer_widgets":"","ocean_shortcode_before_footer_bottom":"","ocean_shortcode_after_footer_bottom":"","ocean_display_top_bar":"default","ocean_display_header":"default","ocean_header_style":"","ocean_center_header_left_menu":"","ocean_custom_header_template":"","ocean_custom_logo":0,"ocean_custom_retina_logo":0,"ocean_custom_logo_max_width":0,"ocean_custom_logo_tablet_max_width":0,"ocean_custom_logo_mobile_max_width":0,"ocean_custom_logo_max_height":0,"ocean_custom_logo_tablet_max_height":0,"ocean_custom_logo_mobile_max_height":0,"ocean_header_custom_menu":"","ocean_menu_typo_font_family":"","ocean_menu_typo_font_subset":"","ocean_menu_typo_font_size":0,"ocean_menu_typo_font_size_tablet":0,"ocean_menu_typo_font_size_mobile":0,"ocean_menu_typo_font_size_unit":"px","ocean_menu_typo_font_weight":"","ocean_menu_typo_font_weight_tablet":"","ocean_menu_typo_font_weight_mobile":"","ocean_menu_typo_transform":"","ocean_menu_typo_transform_tablet":"","ocean_menu_typo_transform_mobile":"","ocean_menu_typo_line_height":0,"ocean_menu_typo_line_height_tablet":0,"ocean_menu_typo_line_height_mobile":0,"ocean_menu_typo_line_height_unit":"","ocean_menu_typo_spacing":0,"ocean_menu_typo_spacing_tablet":0,"ocean_menu_typo_spacing_mobile":0,"ocean_menu_typo_spacing_unit":"","ocean_menu_link_color":"","ocean_menu_link_color_hover":"","ocean_menu_link_color_active":"","ocean_menu_link_background":"","ocean_menu_link_hover_background":"","ocean_menu_link_active_background":"","ocean_menu_social_links_bg":"","ocean_menu_social_hover_links_bg":"","ocean_menu_social_links_color":"","ocean_menu_social_hover_links_color":"","ocean_disable_title":"default","ocean_disable_heading":"default","ocean_post_title":"","ocean_post_subheading":"","ocean_post_title_style":"","ocean_post_title_background_color":"","ocean_post_title_background":0,"ocean_post_title_bg_image_position":"","ocean_post_title_bg_image_attachment":"","ocean_post_title_bg_image_repeat":"","ocean_post_title_bg_image_size":"","ocean_post_title_height":0,"ocean_post_title_bg_overlay":0.5,"ocean_post_title_bg_overlay_color":"","ocean_disable_breadcrumbs":"default","ocean_breadcrumbs_color":"","ocean_breadcrumbs_separator_color":"","ocean_breadcrumbs_links_color":"","ocean_breadcrumbs_links_hover_color":"","ocean_display_footer_widgets":"default","ocean_display_footer_bottom":"default","ocean_custom_footer_template":"","osh_disable_topbar_sticky":"default","osh_disable_header_sticky":"default","osh_sticky_header_style":"default","osh_sticky_header_effect":"","osh_custom_sticky_logo":0,"osh_custom_retina_sticky_logo":0,"osh_custom_sticky_logo_height":0,"osh_background_color":"","osh_links_color":"","osh_links_hover_color":"","osh_links_active_color":"","osh_links_bg_color":"","osh_links_hover_bg_color":"","osh_links_active_bg_color":"","osh_menu_social_links_color":"","osh_menu_social_hover_links_color":"","ocean_post_oembed":"","ocean_post_self_hosted_media":"","ocean_post_video_embed":"","ocean_link_format":"","ocean_link_format_target":"self","ocean_quote_format":"","ocean_quote_format_link":"post","ocean_gallery_link_images":"on","ocean_gallery_id":[],"footnotes":""},"categories":[16,17],"tags":[],"class_list":["post-2159","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-data","entry","has-media"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique - askem<\/title>\n<meta name=\"description\" content=\"ASKEM BUREAU D&#039;\u00c9TUDES ET DE FORMATION NUM\u00c9RIQUE. Nous vous assistons dans la transformation num\u00e9rique de vos outils, services et organisations tout en pla\u00e7ant l\u2019humain au c\u0153ur de notre d\u00e9marche d\u2019accompagnement.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/askem.eu\/en\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique - askem\" \/>\n<meta property=\"og:description\" content=\"ASKEM BUREAU D&#039;\u00c9TUDES ET DE FORMATION NUM\u00c9RIQUE. Nous vous assistons dans la transformation num\u00e9rique de vos outils, services et organisations tout en pla\u00e7ant l\u2019humain au c\u0153ur de notre d\u00e9marche d\u2019accompagnement.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/askem.eu\/en\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/\" \/>\n<meta property=\"og:site_name\" content=\"askem\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/fb.me\/askem.eu\" \/>\n<meta property=\"article:published_time\" content=\"2026-04-01T11:01:43+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-04-01T11:01:49+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2026\/04\/sujet-askem-2026-04-01.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1211\" \/>\n\t<meta property=\"og:image:height\" content=\"1211\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"askemadmin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"askemadmin\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/\"},\"author\":{\"name\":\"askemadmin\",\"@id\":\"https:\\\/\\\/askem.eu\\\/#\\\/schema\\\/person\\\/8bbee74ab9a977d56bf4826662e9d2e9\"},\"headline\":\"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique\",\"datePublished\":\"2026-04-01T11:01:43+00:00\",\"dateModified\":\"2026-04-01T11:01:49+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/\"},\"wordCount\":977,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\/\\/askem.eu\\/wp-content\\/uploads\\/2026\\/04\\/sujet-askem-2026-04-01.png\",\"articleSection\":[\"AI\",\"Data\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/\",\"url\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/\",\"name\":\"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique - askem\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\/\\/askem.eu\\/wp-content\\/uploads\\/2026\\/04\\/sujet-askem-2026-04-01.png\",\"datePublished\":\"2026-04-01T11:01:43+00:00\",\"dateModified\":\"2026-04-01T11:01:49+00:00\",\"description\":\"ASKEM BUREAU D'\u00c9TUDES ET DE FORMATION NUM\u00c9RIQUE. Nous vous assistons dans la transformation num\u00e9rique de vos outils, services et organisations tout en pla\u00e7ant l\u2019humain au c\u0153ur de notre d\u00e9marche d\u2019accompagnement.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#primaryimage\",\"url\":\"https:\\/\\/askem.eu\\/wp-content\\/uploads\\/2026\\/04\\/sujet-askem-2026-04-01.png\",\"contentUrl\":\"https:\\/\\/askem.eu\\/wp-content\\/uploads\\/2026\\/04\\/sujet-askem-2026-04-01.png\",\"width\":1211,\"height\":1211},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/askem.eu\\\/2026\\\/04\\\/01\\\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Accueil\",\"item\":\"https:\\\/\\\/askem.eu\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/askem.eu\\\/#website\",\"url\":\"https:\\\/\\\/askem.eu\\\/\",\"name\":\"askem\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/askem.eu\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/askem.eu\\\/#organization\",\"name\":\"Askem\",\"url\":\"https:\\\/\\\/askem.eu\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/askem.eu\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\/\\/mlpi0fxo3sth.i.optimole.com\\/cb:3obA.c61\\/w:760\\/h:480\\/q:mauto\\/f:best\\/https:\\/\\/askem.eu\\/wp-content\\/uploads\\/2020\\/10\\/logoGalaxieAskem3.png\",\"contentUrl\":\"https:\\/\\/mlpi0fxo3sth.i.optimole.com\\/cb:3obA.c61\\/w:760\\/h:480\\/q:mauto\\/f:best\\/https:\\/\\/askem.eu\\/wp-content\\/uploads\\/2020\\/10\\/logoGalaxieAskem3.png\",\"width\":760,\"height\":480,\"caption\":\"Askem\"},\"image\":{\"@id\":\"https:\\\/\\\/askem.eu\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/fb.me\\\/askem.eu\",\"https:\\\/\\\/linkedin.com\\\/company\\\/askem-eu\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/askem.eu\\\/#\\\/schema\\\/person\\\/8bbee74ab9a977d56bf4826662e9d2e9\",\"name\":\"askemadmin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a202f744ee3a4b6fdbe2ceb57fd84c72559337791a276662270d8d2fb7842e3f?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a202f744ee3a4b6fdbe2ceb57fd84c72559337791a276662270d8d2fb7842e3f?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a202f744ee3a4b6fdbe2ceb57fd84c72559337791a276662270d8d2fb7842e3f?s=96&d=mm&r=g\",\"caption\":\"askemadmin\"},\"sameAs\":[\"https:\\\/\\\/askem.eu\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique - askem","description":"ASKEM BUREAU D'\u00c9TUDES ET DE FORMATION NUM\u00c9RIQUE. Nous vous assistons dans la transformation num\u00e9rique de vos outils, services et organisations tout en pla\u00e7ant l\u2019humain au c\u0153ur de notre d\u00e9marche d\u2019accompagnement.","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:\/\/askem.eu\/en\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/","og_locale":"en_US","og_type":"article","og_title":"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique - askem","og_description":"ASKEM BUREAU D'\u00c9TUDES ET DE FORMATION NUM\u00c9RIQUE. Nous vous assistons dans la transformation num\u00e9rique de vos outils, services et organisations tout en pla\u00e7ant l\u2019humain au c\u0153ur de notre d\u00e9marche d\u2019accompagnement.","og_url":"https:\/\/askem.eu\/en\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/","og_site_name":"askem","article_publisher":"https:\/\/fb.me\/askem.eu","article_published_time":"2026-04-01T11:01:43+00:00","article_modified_time":"2026-04-01T11:01:49+00:00","og_image":[{"width":1211,"height":1211,"url":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2026\/04\/sujet-askem-2026-04-01.png","type":"image\/png"}],"author":"askemadmin","twitter_card":"summary_large_image","twitter_misc":{"Written by":"askemadmin","Est. reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#article","isPartOf":{"@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/"},"author":{"name":"askemadmin","@id":"https:\/\/askem.eu\/#\/schema\/person\/8bbee74ab9a977d56bf4826662e9d2e9"},"headline":"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique","datePublished":"2026-04-01T11:01:43+00:00","dateModified":"2026-04-01T11:01:49+00:00","mainEntityOfPage":{"@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/"},"wordCount":977,"commentCount":0,"publisher":{"@id":"https:\/\/askem.eu\/#organization"},"image":{"@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#primaryimage"},"thumbnailUrl":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2026\/04\/sujet-askem-2026-04-01.png","articleSection":["AI","Data"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/","url":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/","name":"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique - askem","isPartOf":{"@id":"https:\/\/askem.eu\/#website"},"primaryImageOfPage":{"@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#primaryimage"},"image":{"@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#primaryimage"},"thumbnailUrl":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2026\/04\/sujet-askem-2026-04-01.png","datePublished":"2026-04-01T11:01:43+00:00","dateModified":"2026-04-01T11:01:49+00:00","description":"ASKEM BUREAU D'\u00c9TUDES ET DE FORMATION NUM\u00c9RIQUE. Nous vous assistons dans la transformation num\u00e9rique de vos outils, services et organisations tout en pla\u00e7ant l\u2019humain au c\u0153ur de notre d\u00e9marche d\u2019accompagnement.","breadcrumb":{"@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#primaryimage","url":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2026\/04\/sujet-askem-2026-04-01.png","contentUrl":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2026\/04\/sujet-askem-2026-04-01.png","width":1211,"height":1211},{"@type":"BreadcrumbList","@id":"https:\/\/askem.eu\/2026\/04\/01\/qdrant-base-vectorielle-open-source-pour-le-rag-et-la-recherche-semantique\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Accueil","item":"https:\/\/askem.eu\/"},{"@type":"ListItem","position":2,"name":"Qdrant : base vectorielle open source pour le RAG et la recherche s\u00e9mantique"}]},{"@type":"WebSite","@id":"https:\/\/askem.eu\/#website","url":"https:\/\/askem.eu\/","name":"askem","description":"","publisher":{"@id":"https:\/\/askem.eu\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/askem.eu\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/askem.eu\/#organization","name":"Askem","url":"https:\/\/askem.eu\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/askem.eu\/#\/schema\/logo\/image\/","url":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:760\/h:480\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2020\/10\/logoGalaxieAskem3.png","contentUrl":"https:\/\/mlpi0fxo3sth.i.optimole.com\/cb:3obA.c61\/w:760\/h:480\/q:mauto\/f:best\/https:\/\/askem.eu\/wp-content\/uploads\/2020\/10\/logoGalaxieAskem3.png","width":760,"height":480,"caption":"Askem"},"image":{"@id":"https:\/\/askem.eu\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/fb.me\/askem.eu","https:\/\/linkedin.com\/company\/askem-eu"]},{"@type":"Person","@id":"https:\/\/askem.eu\/#\/schema\/person\/8bbee74ab9a977d56bf4826662e9d2e9","name":"askemadmin","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/a202f744ee3a4b6fdbe2ceb57fd84c72559337791a276662270d8d2fb7842e3f?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/a202f744ee3a4b6fdbe2ceb57fd84c72559337791a276662270d8d2fb7842e3f?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/a202f744ee3a4b6fdbe2ceb57fd84c72559337791a276662270d8d2fb7842e3f?s=96&d=mm&r=g","caption":"askemadmin"},"sameAs":["https:\/\/askem.eu"]}]}},"_links":{"self":[{"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/posts\/2159","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/comments?post=2159"}],"version-history":[{"count":1,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/posts\/2159\/revisions"}],"predecessor-version":[{"id":2161,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/posts\/2159\/revisions\/2161"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/media\/2160"}],"wp:attachment":[{"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/media?parent=2159"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/categories?post=2159"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/askem.eu\/en\/wp-json\/wp\/v2\/tags?post=2159"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}