[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-qdrant-qdrant-vertical-scaling":3,"mdc--ypensp-key":47,"related-repo-qdrant-qdrant-vertical-scaling":367,"related-org-qdrant-qdrant-vertical-scaling":466},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":42,"sourceUrl":45,"mdContent":46},"qdrant-vertical-scaling","perform vertical scaling for Qdrant nodes","Guides Qdrant vertical scaling decisions. Use when someone asks 'how to scale up a node', 'need more RAM', 'upgrade node size', 'vertical scaling', 'resize cluster', 'scale up vs scale out', or when memory\u002FCPU is insufficient on current nodes. Also use when someone wants to avoid the complexity of horizontal scaling.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"qdrant","Qdrant","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fqdrant.png",[12,16,17,20],{"name":13,"slug":14,"type":15},"Performance","performance","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Scaling","scaling",{"name":21,"slug":22,"type":15},"Database","database",197,"https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills","2026-07-16T06:02:59.407518",null,23,[29,30,31,32,33,34,35,36,37,14,8,38,19,39,40,41],"agent-skills","ai-agents","claude-code","codex","cursor","embeddings","hybrid-search","monitoring","multitenancy","quantization","search-quality","vector-database","vector-search",{"repoUrl":24,"stars":23,"forks":27,"topics":43,"description":44},[29,30,31,32,33,34,35,36,37,14,8,38,19,39,40,41],"Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, Java","https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fqdrant-scaling\u002Fscaling-data-volume\u002Fvertical-scaling","---\nname: qdrant-vertical-scaling\ndescription: \"Guides Qdrant vertical scaling decisions. Use when someone asks 'how to scale up a node', 'need more RAM', 'upgrade node size', 'vertical scaling', 'resize cluster', 'scale up vs scale out', or when memory\u002FCPU is insufficient on current nodes. Also use when someone wants to avoid the complexity of horizontal scaling.\"\n---\n\n# What to Do When Qdrant Needs to Scale Vertically\n\nVertical scaling means increasing CPU, RAM, or disk on existing nodes rather than adding more nodes. This is the recommended first step before considering horizontal scaling. Vertical scaling is simpler, avoids distributed system complexity, and is reversible.\n\n- Vertical scaling for Qdrant Cloud is done through the [Qdrant Cloud Console](https:\u002F\u002Fcloud.qdrant.io\u002F)\n- For self-hosted deployments, resize the underlying VM or container resources\n\n## When to Scale Vertically\n\nUse when: current node resources (RAM, CPU, disk) are insufficient, but the workload doesn't yet require distribution.\n\n- RAM usage approaching 80% of available memory (OS page cache eviction starts, severe performance degradation)\n- CPU saturation during query serving or indexing\n- Disk space running low for on-disk vectors and payloads\n- A single node can handle up to ~100M vectors depending on dimensions and quantization\n- For non-production workloads, which are tolerant to single-point-of-failure and don't require high availability\n\n\n## How to Scale Vertically in Qdrant Cloud\n\nVertical scaling is managed through the Qdrant Cloud Console.\n\n- Log into [Qdrant Cloud Console](https:\u002F\u002Fcloud.qdrant.io\u002F) or use [CLI tool](https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fqcloud-cli)\n- Select the cluster to resize\n- Choose a larger node configuration (more RAM, CPU, or both)\n- The upgrade process involves a rolling restart with no downtime if replication is configured\n- Ensure `replication_factor: 2` or higher before resizing to maintain availability during the rolling restart\n\n**Important:** Scaling up is straightforward. Scaling down requires care -- if the working set no longer fits in RAM after downsizing, performance will degrade severely due to cache eviction. Always load test before scaling down.\n\n\n## RAM Sizing Guidelines\n\nRAM is the most critical resource for Qdrant performance. Use these guidelines to right-size.\n\n- Exact estimation of RAM usage is difficult; use this simple approximate formula: `num_vectors * dimensions * 4 bytes * 1.5` for full-precision vectors in RAM\n- With scalar quantization: divide by 4 (INT8 reduces each float32 to 1 byte) [Quantization](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F)\n- With binary quantization: divide by 32 [Binary quantization](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F?s=binary-quantization)\n- Add overhead for HNSW index (~20-30% of vector data), payload indexes, and WAL\n- Reserve 20% headroom for optimizer operations and OS cache\n- Monitor actual usage via Grafana\u002FPrometheus before and after resizing [Monitoring](..\u002F..\u002F..\u002Fqdrant-monitoring\u002FSKILL.md)\n\n\n## When Vertical Scaling Is No Longer Enough\n\nRecognize these signals that it's time to go horizontal:\n\n- Data volume exceeds what a single node can hold even with quantization and mmap\n- IOPS are saturated (more nodes = more independent disk I\u002FO)\n- Need fault tolerance (requires replication across nodes)\n- Need tenant isolation via dedicated shards\n- Single-node CPU is maxed and query latency is unacceptable\n- Next vertical scaling step is the largest available node size. You might need to be able to temporarily scale up to the larger node size to do batch operations or recovery. If you are already at the largest node size, you won't be able to do that.\n\nWhen you hit these limits, see [Horizontal Scaling](..\u002Fhorizontal-scaling\u002FSKILL.md) for guidance on sharding and node planning.\n\n\n## What NOT to Do\n\n- Do not scale down RAM without load testing first (cache eviction = severe latency degradation that can last days)\n- Do not ignore the 80% RAM threshold (performance cliff, not gradual degradation)\n- Do not skip replication before resizing in Cloud (rolling restart without replicas = downtime)\n- Do not jump to horizontal scaling before exhausting vertical options (adds permanent operational complexity)\n- Do not assume more CPU always helps (IOPS-bound workloads won't improve with more cores)",{"data":48,"body":49},{"name":4,"description":6},{"type":50,"children":51},"root",[52,61,67,91,98,103,131,137,142,193,204,210,215,276,282,287,320,333,339],{"type":53,"tag":54,"props":55,"children":57},"element","h1",{"id":56},"what-to-do-when-qdrant-needs-to-scale-vertically",[58],{"type":59,"value":60},"text","What to Do When Qdrant Needs to Scale Vertically",{"type":53,"tag":62,"props":63,"children":64},"p",{},[65],{"type":59,"value":66},"Vertical scaling means increasing CPU, RAM, or disk on existing nodes rather than adding more nodes. This is the recommended first step before considering horizontal scaling. Vertical scaling is simpler, avoids distributed system complexity, and is reversible.",{"type":53,"tag":68,"props":69,"children":70},"ul",{},[71,86],{"type":53,"tag":72,"props":73,"children":74},"li",{},[75,77],{"type":59,"value":76},"Vertical scaling for Qdrant Cloud is done through the ",{"type":53,"tag":78,"props":79,"children":83},"a",{"href":80,"rel":81},"https:\u002F\u002Fcloud.qdrant.io\u002F",[82],"nofollow",[84],{"type":59,"value":85},"Qdrant Cloud Console",{"type":53,"tag":72,"props":87,"children":88},{},[89],{"type":59,"value":90},"For self-hosted deployments, resize the underlying VM or container resources",{"type":53,"tag":92,"props":93,"children":95},"h2",{"id":94},"when-to-scale-vertically",[96],{"type":59,"value":97},"When to Scale Vertically",{"type":53,"tag":62,"props":99,"children":100},{},[101],{"type":59,"value":102},"Use when: current node resources (RAM, CPU, disk) are insufficient, but the workload doesn't yet require distribution.",{"type":53,"tag":68,"props":104,"children":105},{},[106,111,116,121,126],{"type":53,"tag":72,"props":107,"children":108},{},[109],{"type":59,"value":110},"RAM usage approaching 80% of available memory (OS page cache eviction starts, severe performance degradation)",{"type":53,"tag":72,"props":112,"children":113},{},[114],{"type":59,"value":115},"CPU saturation during query serving or indexing",{"type":53,"tag":72,"props":117,"children":118},{},[119],{"type":59,"value":120},"Disk space running low for on-disk vectors and payloads",{"type":53,"tag":72,"props":122,"children":123},{},[124],{"type":59,"value":125},"A single node can handle up to ~100M vectors depending on dimensions and quantization",{"type":53,"tag":72,"props":127,"children":128},{},[129],{"type":59,"value":130},"For non-production workloads, which are tolerant to single-point-of-failure and don't require high availability",{"type":53,"tag":92,"props":132,"children":134},{"id":133},"how-to-scale-vertically-in-qdrant-cloud",[135],{"type":59,"value":136},"How to Scale Vertically in Qdrant Cloud",{"type":53,"tag":62,"props":138,"children":139},{},[140],{"type":59,"value":141},"Vertical scaling is managed through the Qdrant Cloud Console.",{"type":53,"tag":68,"props":143,"children":144},{},[145,164,169,174,179],{"type":53,"tag":72,"props":146,"children":147},{},[148,150,155,157],{"type":59,"value":149},"Log into ",{"type":53,"tag":78,"props":151,"children":153},{"href":80,"rel":152},[82],[154],{"type":59,"value":85},{"type":59,"value":156}," or use ",{"type":53,"tag":78,"props":158,"children":161},{"href":159,"rel":160},"https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fqcloud-cli",[82],[162],{"type":59,"value":163},"CLI tool",{"type":53,"tag":72,"props":165,"children":166},{},[167],{"type":59,"value":168},"Select the cluster to resize",{"type":53,"tag":72,"props":170,"children":171},{},[172],{"type":59,"value":173},"Choose a larger node configuration (more RAM, CPU, or both)",{"type":53,"tag":72,"props":175,"children":176},{},[177],{"type":59,"value":178},"The upgrade process involves a rolling restart with no downtime if replication is configured",{"type":53,"tag":72,"props":180,"children":181},{},[182,184,191],{"type":59,"value":183},"Ensure ",{"type":53,"tag":185,"props":186,"children":188},"code",{"className":187},[],[189],{"type":59,"value":190},"replication_factor: 2",{"type":59,"value":192}," or higher before resizing to maintain availability during the rolling restart",{"type":53,"tag":62,"props":194,"children":195},{},[196,202],{"type":53,"tag":197,"props":198,"children":199},"strong",{},[200],{"type":59,"value":201},"Important:",{"type":59,"value":203}," Scaling up is straightforward. Scaling down requires care -- if the working set no longer fits in RAM after downsizing, performance will degrade severely due to cache eviction. Always load test before scaling down.",{"type":53,"tag":92,"props":205,"children":207},{"id":206},"ram-sizing-guidelines",[208],{"type":59,"value":209},"RAM Sizing Guidelines",{"type":53,"tag":62,"props":211,"children":212},{},[213],{"type":59,"value":214},"RAM is the most critical resource for Qdrant performance. Use these guidelines to right-size.",{"type":53,"tag":68,"props":216,"children":217},{},[218,231,243,255,260,265],{"type":53,"tag":72,"props":219,"children":220},{},[221,223,229],{"type":59,"value":222},"Exact estimation of RAM usage is difficult; use this simple approximate formula: ",{"type":53,"tag":185,"props":224,"children":226},{"className":225},[],[227],{"type":59,"value":228},"num_vectors * dimensions * 4 bytes * 1.5",{"type":59,"value":230}," for full-precision vectors in RAM",{"type":53,"tag":72,"props":232,"children":233},{},[234,236],{"type":59,"value":235},"With scalar quantization: divide by 4 (INT8 reduces each float32 to 1 byte) ",{"type":53,"tag":78,"props":237,"children":240},{"href":238,"rel":239},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F",[82],[241],{"type":59,"value":242},"Quantization",{"type":53,"tag":72,"props":244,"children":245},{},[246,248],{"type":59,"value":247},"With binary quantization: divide by 32 ",{"type":53,"tag":78,"props":249,"children":252},{"href":250,"rel":251},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F?s=binary-quantization",[82],[253],{"type":59,"value":254},"Binary quantization",{"type":53,"tag":72,"props":256,"children":257},{},[258],{"type":59,"value":259},"Add overhead for HNSW index (~20-30% of vector data), payload indexes, and WAL",{"type":53,"tag":72,"props":261,"children":262},{},[263],{"type":59,"value":264},"Reserve 20% headroom for optimizer operations and OS cache",{"type":53,"tag":72,"props":266,"children":267},{},[268,270],{"type":59,"value":269},"Monitor actual usage via Grafana\u002FPrometheus before and after resizing ",{"type":53,"tag":78,"props":271,"children":273},{"href":272},"..\u002F..\u002F..\u002Fqdrant-monitoring\u002FSKILL.md",[274],{"type":59,"value":275},"Monitoring",{"type":53,"tag":92,"props":277,"children":279},{"id":278},"when-vertical-scaling-is-no-longer-enough",[280],{"type":59,"value":281},"When Vertical Scaling Is No Longer Enough",{"type":53,"tag":62,"props":283,"children":284},{},[285],{"type":59,"value":286},"Recognize these signals that it's time to go horizontal:",{"type":53,"tag":68,"props":288,"children":289},{},[290,295,300,305,310,315],{"type":53,"tag":72,"props":291,"children":292},{},[293],{"type":59,"value":294},"Data volume exceeds what a single node can hold even with quantization and mmap",{"type":53,"tag":72,"props":296,"children":297},{},[298],{"type":59,"value":299},"IOPS are saturated (more nodes = more independent disk I\u002FO)",{"type":53,"tag":72,"props":301,"children":302},{},[303],{"type":59,"value":304},"Need fault tolerance (requires replication across nodes)",{"type":53,"tag":72,"props":306,"children":307},{},[308],{"type":59,"value":309},"Need tenant isolation via dedicated shards",{"type":53,"tag":72,"props":311,"children":312},{},[313],{"type":59,"value":314},"Single-node CPU is maxed and query latency is unacceptable",{"type":53,"tag":72,"props":316,"children":317},{},[318],{"type":59,"value":319},"Next vertical scaling step is the largest available node size. You might need to be able to temporarily scale up to the larger node size to do batch operations or recovery. If you are already at the largest node size, you won't be able to do that.",{"type":53,"tag":62,"props":321,"children":322},{},[323,325,331],{"type":59,"value":324},"When you hit these limits, see ",{"type":53,"tag":78,"props":326,"children":328},{"href":327},"..\u002Fhorizontal-scaling\u002FSKILL.md",[329],{"type":59,"value":330},"Horizontal Scaling",{"type":59,"value":332}," for guidance on sharding and node planning.",{"type":53,"tag":92,"props":334,"children":336},{"id":335},"what-not-to-do",[337],{"type":59,"value":338},"What NOT to Do",{"type":53,"tag":68,"props":340,"children":341},{},[342,347,352,357,362],{"type":53,"tag":72,"props":343,"children":344},{},[345],{"type":59,"value":346},"Do not scale down RAM without load testing first (cache eviction = severe latency degradation that can last days)",{"type":53,"tag":72,"props":348,"children":349},{},[350],{"type":59,"value":351},"Do not ignore the 80% RAM threshold (performance cliff, not gradual degradation)",{"type":53,"tag":72,"props":353,"children":354},{},[355],{"type":59,"value":356},"Do not skip replication before resizing in Cloud (rolling restart without replicas = downtime)",{"type":53,"tag":72,"props":358,"children":359},{},[360],{"type":59,"value":361},"Do not jump to horizontal scaling before exhausting vertical options (adds permanent operational complexity)",{"type":53,"tag":72,"props":363,"children":364},{},[365],{"type":59,"value":366},"Do not assume more CPU always helps (IOPS-bound workloads won't improve with more cores)",{"items":368,"total":465},[369,384,399,414,429,442,455],{"slug":370,"name":370,"fn":371,"description":372,"org":373,"tags":374,"stars":23,"repoUrl":24,"updatedAt":383},"qdrant-advisor","diagnose and troubleshoot Qdrant deployments","Diagnose, troubleshoot, and advise on any Qdrant deployment by loading the latest official Qdrant skills live from skills.qdrant.tech. Use this whenever someone raises a Qdrant problem or question — slow or degraded search, high or growing memory \u002F OOM crashes, optimizer stuck or slow, indexing slowness, scaling and sharding decisions (node count, QPS, latency, multitenancy, vertical vs horizontal), poor or irrelevant search results, hybrid search and reranking, embedding-model migration, version upgrades and compatibility, monitoring and observability (Prometheus, Grafana, health checks, \u002Fmetrics, \u002Ftelemetry), deployment choices (local, Docker, self-hosted, Qdrant Cloud, embedded), or client-SDK questions (Python, TypeScript, Rust, Go, .NET, Java). Trigger especially when the context is clearly a Qdrant cluster, collection, or vector-search deployment. Always prefer this skill over answering from memory: it pulls current, authoritative guidance and only the relevant context.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[375,376,379,382],{"name":21,"slug":22,"type":15},{"name":377,"slug":378,"type":15},"Debugging","debugging",{"name":380,"slug":381,"type":15},"Operations","operations",{"name":9,"slug":8,"type":15},"2026-07-16T06:02:56.675293",{"slug":385,"name":385,"fn":386,"description":387,"org":388,"tags":389,"stars":23,"repoUrl":24,"updatedAt":398},"qdrant-clients-sdk","integrate Qdrant client SDKs","Qdrant provides client SDKs for various programming languages, allowing easy integration with Qdrant deployments.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[390,393,394,395],{"name":391,"slug":392,"type":15},"API Development","api-development",{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":396,"slug":397,"type":15},"SDK","sdk","2026-07-16T05:59:57.799789",{"slug":400,"name":400,"fn":401,"description":402,"org":403,"tags":404,"stars":23,"repoUrl":24,"updatedAt":413},"qdrant-deployment-options","select Qdrant deployment options","Guides Qdrant deployment selection. Use when someone asks 'how to deploy Qdrant', 'Docker vs Cloud', 'local mode', 'embedded Qdrant', 'Qdrant EDGE', 'which deployment option', 'self-hosted vs cloud', or 'need lowest latency deployment'. Also use when choosing between deployment types for a new project.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[405,406,409,412],{"name":21,"slug":22,"type":15},{"name":407,"slug":408,"type":15},"Deployment","deployment",{"name":410,"slug":411,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-19T05:38:33.324265",{"slug":415,"name":415,"fn":416,"description":417,"org":418,"tags":419,"stars":23,"repoUrl":24,"updatedAt":428},"qdrant-edge","build applications with Qdrant Edge","Guides building on Qdrant Edge, the embedded in-process shard. Use when someone asks 'how to sync Edge with the server', 'keep a local shard in sync with Qdrant Cloud', 'BM25 or keyword search on Edge', 'hybrid search on Edge', 'embeddings on device', 'Edge snapshots', 'apply a partial snapshot', 'why is my Edge search empty after inserts', or is writing custom sync, BM25, or fusion code against qdrant-edge. Also use when deciding what Edge ships built-in versus what you must implement.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[420,421,424,425],{"name":21,"slug":22,"type":15},{"name":422,"slug":423,"type":15},"Edge","edge",{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},"Search","search","2026-07-16T06:01:19.179105",{"slug":430,"name":430,"fn":431,"description":432,"org":433,"tags":434,"stars":23,"repoUrl":24,"updatedAt":441},"qdrant-horizontal-scaling","guide Qdrant horizontal scaling decisions","Diagnoses and guides Qdrant horizontal scaling decisions. Use when someone asks 'vertical or horizontal?', 'how many nodes?', 'how many shards?', 'how to add nodes', 'resharding', 'data doesn't fit', or 'need more capacity'. Also use when data growth outpaces current deployment.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[435,438,439,440],{"name":436,"slug":437,"type":15},"Architecture","architecture",{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"2026-07-19T05:38:32.348551",{"slug":443,"name":443,"fn":444,"description":445,"org":446,"tags":447,"stars":23,"repoUrl":24,"updatedAt":454},"qdrant-hybrid-search","implement hybrid search in Qdrant","Explains hybrid search in Qdrant. Use when someone asks 'how do I setup hybrid search?', 'how to combine keyword and semantic search?', 'sparse plus dense vectors?', 'missing keyword matches', 'how to combine results from multiple searches?' and 'combining multiple representations'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[448,451,452,453],{"name":449,"slug":450,"type":15},"AI Infrastructure","ai-infrastructure",{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},"2026-07-16T06:00:12.328122",{"slug":456,"name":456,"fn":457,"description":458,"org":459,"tags":460,"stars":23,"repoUrl":24,"updatedAt":464},"qdrant-hybrid-search-combining","combine hybrid search scores in Qdrant","Fusing scores from multiple searches into a single ranked result (RRF, DBSF, custom fusion). Use when someone asks 'RRF or DBSF?', 'how to combine sparse and dense', 'how to combine scores from multiple searches?', 'custom fusion', or 'fusion is not producing good results'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[461,462,463],{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},"2026-07-16T06:01:49.401142",30,{"items":467,"total":465},[468,475,482,489,496,503,510,516,527,538,548,559],{"slug":370,"name":370,"fn":371,"description":372,"org":469,"tags":470,"stars":23,"repoUrl":24,"updatedAt":383},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[471,472,473,474],{"name":21,"slug":22,"type":15},{"name":377,"slug":378,"type":15},{"name":380,"slug":381,"type":15},{"name":9,"slug":8,"type":15},{"slug":385,"name":385,"fn":386,"description":387,"org":476,"tags":477,"stars":23,"repoUrl":24,"updatedAt":398},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[478,479,480,481],{"name":391,"slug":392,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":396,"slug":397,"type":15},{"slug":400,"name":400,"fn":401,"description":402,"org":483,"tags":484,"stars":23,"repoUrl":24,"updatedAt":413},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[485,486,487,488],{"name":21,"slug":22,"type":15},{"name":407,"slug":408,"type":15},{"name":410,"slug":411,"type":15},{"name":9,"slug":8,"type":15},{"slug":415,"name":415,"fn":416,"description":417,"org":490,"tags":491,"stars":23,"repoUrl":24,"updatedAt":428},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[492,493,494,495],{"name":21,"slug":22,"type":15},{"name":422,"slug":423,"type":15},{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},{"slug":430,"name":430,"fn":431,"description":432,"org":497,"tags":498,"stars":23,"repoUrl":24,"updatedAt":441},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[499,500,501,502],{"name":436,"slug":437,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},{"slug":443,"name":443,"fn":444,"description":445,"org":504,"tags":505,"stars":23,"repoUrl":24,"updatedAt":454},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[506,507,508,509],{"name":449,"slug":450,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},{"slug":456,"name":456,"fn":457,"description":458,"org":511,"tags":512,"stars":23,"repoUrl":24,"updatedAt":464},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[513,514,515],{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},{"slug":517,"name":517,"fn":518,"description":519,"org":520,"tags":521,"stars":23,"repoUrl":24,"updatedAt":526},"qdrant-hybrid-search-prefetches","configure hybrid search with prefetch queries","Constructing prefetch queries for hybrid retrieval, including sparse\u002Fdense and multi-field setups, and choosing a sparse embedding model. Use when someone asks 'dense and sparse in one search?', 'how to combine multiple fields for retrieval?', 'payloads or sparse vectors for lexical?', 'which sparse embedding model to use?', or 'BM25 vs SPLADE?'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[522,523,524,525],{"name":391,"slug":392,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":426,"slug":427,"type":15},"2026-07-16T06:03:02.305907",{"slug":528,"name":528,"fn":529,"description":530,"org":531,"tags":532,"stars":23,"repoUrl":24,"updatedAt":537},"qdrant-indexing-performance-optimization","optimize Qdrant indexing and data ingestion","Diagnoses and fixes slow Qdrant indexing and data ingestion. Use when someone reports 'uploads are slow', 'indexing takes forever', 'optimizer is stuck', 'HNSW build time too long', or 'data uploaded but search is bad'. Also use when optimizer status shows errors, segments won't merge, or indexing threshold questions arise.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[533,534,535,536],{"name":21,"slug":22,"type":15},{"name":377,"slug":378,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-19T05:38:36.321999",{"slug":539,"name":539,"fn":540,"description":541,"org":542,"tags":543,"stars":23,"repoUrl":24,"updatedAt":547},"qdrant-memory-usage-optimization","optimize Qdrant memory usage","Diagnoses and reduces Qdrant memory usage. Use when someone reports 'memory too high', 'RAM keeps growing', 'node crashed', 'out of memory', 'memory leak', or asks 'why is memory usage so high?', 'how to reduce RAM?'. Also use when memory doesn't match calculations, quantization didn't help, or nodes crash during recovery.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[544,545,546],{"name":377,"slug":378,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T06:00:15.357243",{"slug":549,"name":549,"fn":550,"description":551,"org":552,"tags":553,"stars":23,"repoUrl":24,"updatedAt":558},"qdrant-minimize-latency","optimize Qdrant query latency","Guides Qdrant query latency optimization. Use when someone asks 'search is slow', 'how to reduce latency', 'p99 is too high', 'tail latency', 'single query too slow', 'how to make search faster', or 'latency spikes'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[554,555,556,557],{"name":21,"slug":22,"type":15},{"name":377,"slug":378,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T05:59:17.325839",{"slug":560,"name":560,"fn":561,"description":562,"org":563,"tags":564,"stars":23,"repoUrl":24,"updatedAt":571},"qdrant-model-migration","migrate embedding models in Qdrant","Guides embedding model migration in Qdrant without downtime. Use when someone asks 'how to switch embedding models', 'how to migrate vectors', 'how to update to a new model', 'zero-downtime model change', 'how to re-embed my data', or 'can I use two models at once'. Also use when upgrading model dimensions, switching providers, or A\u002FB testing models.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[565,566,569,570],{"name":21,"slug":22,"type":15},{"name":567,"slug":568,"type":15},"Migration","migration",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T06:00:48.841055"]