[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-qdrant-qdrant-search-quality-diagnosis":3,"mdc--1h2wtx-key":49,"related-org-qdrant-qdrant-search-quality-diagnosis":475,"related-repo-qdrant-qdrant-search-quality-diagnosis":630},{"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":44,"sourceUrl":47,"mdContent":48},"qdrant-search-quality-diagnosis","diagnose Qdrant search quality issues","Diagnoses Qdrant search quality issues. Use when someone reports 'results are bad', 'wrong results', 'not relevant results', 'missing matches', 'recall is low', 'approximate search worse than exact', 'which embedding model', 'quality dropped after quantization', 'how to measure retrieval quality', 'build a golden set', 'ground truth dataset', or 'how to score recall@k'. Also use when search quality degrades without obvious changes.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"qdrant","Qdrant","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fqdrant.png",[12,14,17,20],{"name":9,"slug":8,"type":13},"tag",{"name":15,"slug":16,"type":13},"Data Analysis","data-analysis",{"name":18,"slug":19,"type":13},"Search","search",{"name":21,"slug":22,"type":13},"Debugging","debugging",197,"https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills","2026-07-16T06:00:01.549139",null,23,[29,30,31,32,33,34,35,36,37,38,8,39,40,41,42,43],"agent-skills","ai-agents","claude-code","codex","cursor","embeddings","hybrid-search","monitoring","multitenancy","performance","quantization","scaling","search-quality","vector-database","vector-search",{"repoUrl":24,"stars":23,"forks":27,"topics":45,"description":46},[29,30,31,32,33,34,35,36,37,38,8,39,40,41,42,43],"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-search-quality\u002Fdiagnosis","---\nname: qdrant-search-quality-diagnosis\ndescription: \"Diagnoses Qdrant search quality issues. Use when someone reports 'results are bad', 'wrong results', 'not relevant results', 'missing matches', 'recall is low', 'approximate search worse than exact', 'which embedding model', 'quality dropped after quantization', 'how to measure retrieval quality', 'build a golden set', 'ground truth dataset', or 'how to score recall@k'. Also use when search quality degrades without obvious changes.\"\n---\n\n# How to Diagnose Bad Search Quality\n\nBefore tuning, establish baselines. Use exact KNN as ground truth, compare against approximate HNSW. Target >95% recall@K for production.\n\n## Don't Know What's Wrong Yet\n\nUse when: results are irrelevant or missing expected matches and you need to isolate the cause.\n\n- For a no-code quick check, use the Web UI's ANN Recall tab to compare approximate vs exact `recall@k` [Web UI ANN Recall](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Ftutorials-search-engineering\u002Fann-recall\u002F?s=measure-ann-recall-with-the-web-ui)\n- For the same comparison in code (CI gating, regression tests), run each query twice — once approximate, once with `exact=true` — and compute `recall@k` from the overlap [ANN recall in CI](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Ftutorials-search-engineering\u002Fann-recall\u002F?s=automate-in-ci-with-python)\n- Exact search bad = model or search pipeline problem. Exact good, approximate bad = tune HNSW.\n- Check if quantization degrades quality (compare with and without)\n- Check if filters are too restrictive (then you might need to use ACORN)\n- If duplicate results from chunked documents, use Grouping API to deduplicate [Grouping](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=grouping-api)\n\nPayload filtering and sparse vector search are different things. Metadata (dates, categories, tags) goes in payload for filtering. Text content goes in sparse vectors for search.\n\n## Approximate Search Worse Than Exact\n\nUse when: exact search returns good results but HNSW approximation misses them.\n\n- Increase `hnsw_ef` at query time [Search params](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fops-optimization\u002Foptimize\u002F?s=fine-tuning-search-parameters)\n- Increase `ef_construct` (200+ for high quality) [HNSW config](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index)\n- Increase `m` (16 default, 32 for high recall) [HNSW config](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index)\n- Enable oversampling + rescore with quantization [Search with quantization](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F?s=searching-with-quantization)\n- ACORN for filtered queries (v1.16+) [ACORN](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=acorn-search-algorithm)\n\nBinary quantization requires rescore. Without it, quality loss is severe. Use oversampling (3-5x minimum for binary) to recover recall. Always test quantization impact on your data before production. [Quantization](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F)\n\n## Wrong Embedding Model\n\nUse when: exact search also returns bad results.\n\nCheck [Qdrant team recommendations on how to choose an embedding model](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Farticles\u002Fhow-to-choose-an-embedding-model\u002F).\n\nTest top 3 MTEB models on 100-1000 sample queries [Hosted Qdrant inference](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Finference\u002F). Score them against a labeled set to compare apples to apples [Measuring Retrieval Relevance](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fretrieval-relevance\u002F).\n\n## Unoptimized Search Pipeline\n\nUse when: exact search also returns bad results and model choice is confirmed by user.\n\nOptimize search according to advanced search-strategies skill.\n\n## Need a Labeled Baseline to Score Recall, MRR, or NDCG\n\nUse when: user has no golden set, asks \"how do I know if my search is good?\", or needs to gate releases on a retrieval metric.\n\n- Build a labeled query set — human, log-based, or LLM-synthetic — and score retrieval with `ranx` [Measuring Retrieval Relevance](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fretrieval-relevance\u002F)\n- Pick the metric by usage: `Recall@k` for RAG, `MRR`\u002F`Hits@1` for single-answer, `NDCG@k` for re-ranking [Choosing the metric](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fretrieval-relevance\u002F?s=choosing-the-right-metric)\n- For full RAG pipelines, also score generation with Ragas and use the retrieval-vs-generation 2x2 to isolate regressions [Pipeline Output Quality](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fpipeline-output-quality\u002F)\n- Gate CI on a per-metric threshold to catch regressions from embedding-model swaps, prompt changes, or index config changes\n\n## What NOT to Do\n\n- Tune Qdrant before verifying the model is right for the task (most quality issues are model issues)\n- Use binary quantization without rescore (severe quality loss)\n- Set `hnsw_ef` lower than results requested (guaranteed bad recall)\n- Skip payload indexes on filtered fields then blame quality (HNSW can't traverse filtered-out nodes, and filterable HNSW is built only if payload indexes were set up prior)\n- Deploy without baseline recall or other search relevance metrics (no way to measure regressions)\n- Confuse payload filtering with sparse vector search (different things, different config)\n",{"data":50,"body":51},{"name":4,"description":6},{"type":52,"children":53},"root",[54,63,69,76,81,163,168,174,179,262,274,280,285,299,321,327,332,337,343,348,429,435],{"type":55,"tag":56,"props":57,"children":59},"element","h1",{"id":58},"how-to-diagnose-bad-search-quality",[60],{"type":61,"value":62},"text","How to Diagnose Bad Search Quality",{"type":55,"tag":64,"props":65,"children":66},"p",{},[67],{"type":61,"value":68},"Before tuning, establish baselines. Use exact KNN as ground truth, compare against approximate HNSW. Target >95% recall@K for production.",{"type":55,"tag":70,"props":71,"children":73},"h2",{"id":72},"dont-know-whats-wrong-yet",[74],{"type":61,"value":75},"Don't Know What's Wrong Yet",{"type":55,"tag":64,"props":77,"children":78},{},[79],{"type":61,"value":80},"Use when: results are irrelevant or missing expected matches and you need to isolate the cause.",{"type":55,"tag":82,"props":83,"children":84},"ul",{},[85,109,136,141,146,151],{"type":55,"tag":86,"props":87,"children":88},"li",{},[89,91,98,100],{"type":61,"value":90},"For a no-code quick check, use the Web UI's ANN Recall tab to compare approximate vs exact ",{"type":55,"tag":92,"props":93,"children":95},"code",{"className":94},[],[96],{"type":61,"value":97},"recall@k",{"type":61,"value":99}," ",{"type":55,"tag":101,"props":102,"children":106},"a",{"href":103,"rel":104},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Ftutorials-search-engineering\u002Fann-recall\u002F?s=measure-ann-recall-with-the-web-ui",[105],"nofollow",[107],{"type":61,"value":108},"Web UI ANN Recall",{"type":55,"tag":86,"props":110,"children":111},{},[112,114,120,122,127,129],{"type":61,"value":113},"For the same comparison in code (CI gating, regression tests), run each query twice — once approximate, once with ",{"type":55,"tag":92,"props":115,"children":117},{"className":116},[],[118],{"type":61,"value":119},"exact=true",{"type":61,"value":121}," — and compute ",{"type":55,"tag":92,"props":123,"children":125},{"className":124},[],[126],{"type":61,"value":97},{"type":61,"value":128}," from the overlap ",{"type":55,"tag":101,"props":130,"children":133},{"href":131,"rel":132},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Ftutorials-search-engineering\u002Fann-recall\u002F?s=automate-in-ci-with-python",[105],[134],{"type":61,"value":135},"ANN recall in CI",{"type":55,"tag":86,"props":137,"children":138},{},[139],{"type":61,"value":140},"Exact search bad = model or search pipeline problem. Exact good, approximate bad = tune HNSW.",{"type":55,"tag":86,"props":142,"children":143},{},[144],{"type":61,"value":145},"Check if quantization degrades quality (compare with and without)",{"type":55,"tag":86,"props":147,"children":148},{},[149],{"type":61,"value":150},"Check if filters are too restrictive (then you might need to use ACORN)",{"type":55,"tag":86,"props":152,"children":153},{},[154,156],{"type":61,"value":155},"If duplicate results from chunked documents, use Grouping API to deduplicate ",{"type":55,"tag":101,"props":157,"children":160},{"href":158,"rel":159},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=grouping-api",[105],[161],{"type":61,"value":162},"Grouping",{"type":55,"tag":64,"props":164,"children":165},{},[166],{"type":61,"value":167},"Payload filtering and sparse vector search are different things. Metadata (dates, categories, tags) goes in payload for filtering. Text content goes in sparse vectors for search.",{"type":55,"tag":70,"props":169,"children":171},{"id":170},"approximate-search-worse-than-exact",[172],{"type":61,"value":173},"Approximate Search Worse Than Exact",{"type":55,"tag":64,"props":175,"children":176},{},[177],{"type":61,"value":178},"Use when: exact search returns good results but HNSW approximation misses them.",{"type":55,"tag":82,"props":180,"children":181},{},[182,202,221,238,250],{"type":55,"tag":86,"props":183,"children":184},{},[185,187,193,195],{"type":61,"value":186},"Increase ",{"type":55,"tag":92,"props":188,"children":190},{"className":189},[],[191],{"type":61,"value":192},"hnsw_ef",{"type":61,"value":194}," at query time ",{"type":55,"tag":101,"props":196,"children":199},{"href":197,"rel":198},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fops-optimization\u002Foptimize\u002F?s=fine-tuning-search-parameters",[105],[200],{"type":61,"value":201},"Search params",{"type":55,"tag":86,"props":203,"children":204},{},[205,206,212,214],{"type":61,"value":186},{"type":55,"tag":92,"props":207,"children":209},{"className":208},[],[210],{"type":61,"value":211},"ef_construct",{"type":61,"value":213}," (200+ for high quality) ",{"type":55,"tag":101,"props":215,"children":218},{"href":216,"rel":217},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index",[105],[219],{"type":61,"value":220},"HNSW config",{"type":55,"tag":86,"props":222,"children":223},{},[224,225,231,233],{"type":61,"value":186},{"type":55,"tag":92,"props":226,"children":228},{"className":227},[],[229],{"type":61,"value":230},"m",{"type":61,"value":232}," (16 default, 32 for high recall) ",{"type":55,"tag":101,"props":234,"children":236},{"href":216,"rel":235},[105],[237],{"type":61,"value":220},{"type":55,"tag":86,"props":239,"children":240},{},[241,243],{"type":61,"value":242},"Enable oversampling + rescore with quantization ",{"type":55,"tag":101,"props":244,"children":247},{"href":245,"rel":246},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F?s=searching-with-quantization",[105],[248],{"type":61,"value":249},"Search with quantization",{"type":55,"tag":86,"props":251,"children":252},{},[253,255],{"type":61,"value":254},"ACORN for filtered queries (v1.16+) ",{"type":55,"tag":101,"props":256,"children":259},{"href":257,"rel":258},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=acorn-search-algorithm",[105],[260],{"type":61,"value":261},"ACORN",{"type":55,"tag":64,"props":263,"children":264},{},[265,267],{"type":61,"value":266},"Binary quantization requires rescore. Without it, quality loss is severe. Use oversampling (3-5x minimum for binary) to recover recall. Always test quantization impact on your data before production. ",{"type":55,"tag":101,"props":268,"children":271},{"href":269,"rel":270},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F",[105],[272],{"type":61,"value":273},"Quantization",{"type":55,"tag":70,"props":275,"children":277},{"id":276},"wrong-embedding-model",[278],{"type":61,"value":279},"Wrong Embedding Model",{"type":55,"tag":64,"props":281,"children":282},{},[283],{"type":61,"value":284},"Use when: exact search also returns bad results.",{"type":55,"tag":64,"props":286,"children":287},{},[288,290,297],{"type":61,"value":289},"Check ",{"type":55,"tag":101,"props":291,"children":294},{"href":292,"rel":293},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Farticles\u002Fhow-to-choose-an-embedding-model\u002F",[105],[295],{"type":61,"value":296},"Qdrant team recommendations on how to choose an embedding model",{"type":61,"value":298},".",{"type":55,"tag":64,"props":300,"children":301},{},[302,304,311,313,320],{"type":61,"value":303},"Test top 3 MTEB models on 100-1000 sample queries ",{"type":55,"tag":101,"props":305,"children":308},{"href":306,"rel":307},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Finference\u002F",[105],[309],{"type":61,"value":310},"Hosted Qdrant inference",{"type":61,"value":312},". Score them against a labeled set to compare apples to apples ",{"type":55,"tag":101,"props":314,"children":317},{"href":315,"rel":316},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fretrieval-relevance\u002F",[105],[318],{"type":61,"value":319},"Measuring Retrieval Relevance",{"type":61,"value":298},{"type":55,"tag":70,"props":322,"children":324},{"id":323},"unoptimized-search-pipeline",[325],{"type":61,"value":326},"Unoptimized Search Pipeline",{"type":55,"tag":64,"props":328,"children":329},{},[330],{"type":61,"value":331},"Use when: exact search also returns bad results and model choice is confirmed by user.",{"type":55,"tag":64,"props":333,"children":334},{},[335],{"type":61,"value":336},"Optimize search according to advanced search-strategies skill.",{"type":55,"tag":70,"props":338,"children":340},{"id":339},"need-a-labeled-baseline-to-score-recall-mrr-or-ndcg",[341],{"type":61,"value":342},"Need a Labeled Baseline to Score Recall, MRR, or NDCG",{"type":55,"tag":64,"props":344,"children":345},{},[346],{"type":61,"value":347},"Use when: user has no golden set, asks \"how do I know if my search is good?\", or needs to gate releases on a retrieval metric.",{"type":55,"tag":82,"props":349,"children":350},{},[351,368,412,424],{"type":55,"tag":86,"props":352,"children":353},{},[354,356,362,363],{"type":61,"value":355},"Build a labeled query set — human, log-based, or LLM-synthetic — and score retrieval with ",{"type":55,"tag":92,"props":357,"children":359},{"className":358},[],[360],{"type":61,"value":361},"ranx",{"type":61,"value":99},{"type":55,"tag":101,"props":364,"children":366},{"href":315,"rel":365},[105],[367],{"type":61,"value":319},{"type":55,"tag":86,"props":369,"children":370},{},[371,373,379,381,387,389,395,397,403,405],{"type":61,"value":372},"Pick the metric by usage: ",{"type":55,"tag":92,"props":374,"children":376},{"className":375},[],[377],{"type":61,"value":378},"Recall@k",{"type":61,"value":380}," for RAG, ",{"type":55,"tag":92,"props":382,"children":384},{"className":383},[],[385],{"type":61,"value":386},"MRR",{"type":61,"value":388},"\u002F",{"type":55,"tag":92,"props":390,"children":392},{"className":391},[],[393],{"type":61,"value":394},"Hits@1",{"type":61,"value":396}," for single-answer, ",{"type":55,"tag":92,"props":398,"children":400},{"className":399},[],[401],{"type":61,"value":402},"NDCG@k",{"type":61,"value":404}," for re-ranking ",{"type":55,"tag":101,"props":406,"children":409},{"href":407,"rel":408},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fretrieval-relevance\u002F?s=choosing-the-right-metric",[105],[410],{"type":61,"value":411},"Choosing the metric",{"type":55,"tag":86,"props":413,"children":414},{},[415,417],{"type":61,"value":416},"For full RAG pipelines, also score generation with Ragas and use the retrieval-vs-generation 2x2 to isolate regressions ",{"type":55,"tag":101,"props":418,"children":421},{"href":419,"rel":420},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fimprove-search\u002Fpipeline-output-quality\u002F",[105],[422],{"type":61,"value":423},"Pipeline Output Quality",{"type":55,"tag":86,"props":425,"children":426},{},[427],{"type":61,"value":428},"Gate CI on a per-metric threshold to catch regressions from embedding-model swaps, prompt changes, or index config changes",{"type":55,"tag":70,"props":430,"children":432},{"id":431},"what-not-to-do",[433],{"type":61,"value":434},"What NOT to Do",{"type":55,"tag":82,"props":436,"children":437},{},[438,443,448,460,465,470],{"type":55,"tag":86,"props":439,"children":440},{},[441],{"type":61,"value":442},"Tune Qdrant before verifying the model is right for the task (most quality issues are model issues)",{"type":55,"tag":86,"props":444,"children":445},{},[446],{"type":61,"value":447},"Use binary quantization without rescore (severe quality loss)",{"type":55,"tag":86,"props":449,"children":450},{},[451,453,458],{"type":61,"value":452},"Set ",{"type":55,"tag":92,"props":454,"children":456},{"className":455},[],[457],{"type":61,"value":192},{"type":61,"value":459}," lower than results requested (guaranteed bad recall)",{"type":55,"tag":86,"props":461,"children":462},{},[463],{"type":61,"value":464},"Skip payload indexes on filtered fields then blame quality (HNSW can't traverse filtered-out nodes, and filterable HNSW is built only if payload indexes were set up prior)",{"type":55,"tag":86,"props":466,"children":467},{},[468],{"type":61,"value":469},"Deploy without baseline recall or other search relevance metrics (no way to measure regressions)",{"type":55,"tag":86,"props":471,"children":472},{},[473],{"type":61,"value":474},"Confuse payload filtering with sparse vector search (different things, different config)",{"items":476,"total":629},[477,492,507,522,535,549,562,572,583,595,605,616],{"slug":478,"name":478,"fn":479,"description":480,"org":481,"tags":482,"stars":23,"repoUrl":24,"updatedAt":491},"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},[483,486,487,490],{"name":484,"slug":485,"type":13},"Database","database",{"name":21,"slug":22,"type":13},{"name":488,"slug":489,"type":13},"Operations","operations",{"name":9,"slug":8,"type":13},"2026-07-16T06:02:56.675293",{"slug":493,"name":493,"fn":494,"description":495,"org":496,"tags":497,"stars":23,"repoUrl":24,"updatedAt":506},"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},[498,501,502,503],{"name":499,"slug":500,"type":13},"API Development","api-development",{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":504,"slug":505,"type":13},"SDK","sdk","2026-07-16T05:59:57.799789",{"slug":508,"name":508,"fn":509,"description":510,"org":511,"tags":512,"stars":23,"repoUrl":24,"updatedAt":521},"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},[513,514,517,520],{"name":484,"slug":485,"type":13},{"name":515,"slug":516,"type":13},"Deployment","deployment",{"name":518,"slug":519,"type":13},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":13},"2026-07-19T05:38:33.324265",{"slug":523,"name":523,"fn":524,"description":525,"org":526,"tags":527,"stars":23,"repoUrl":24,"updatedAt":534},"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},[528,529,532,533],{"name":484,"slug":485,"type":13},{"name":530,"slug":531,"type":13},"Edge","edge",{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13},"2026-07-16T06:01:19.179105",{"slug":536,"name":536,"fn":537,"description":538,"org":539,"tags":540,"stars":23,"repoUrl":24,"updatedAt":548},"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},[541,544,545,546],{"name":542,"slug":543,"type":13},"Architecture","architecture",{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":547,"slug":40,"type":13},"Scaling","2026-07-19T05:38:32.348551",{"slug":550,"name":550,"fn":551,"description":552,"org":553,"tags":554,"stars":23,"repoUrl":24,"updatedAt":561},"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},[555,558,559,560],{"name":556,"slug":557,"type":13},"AI Infrastructure","ai-infrastructure",{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13},"2026-07-16T06:00:12.328122",{"slug":563,"name":563,"fn":564,"description":565,"org":566,"tags":567,"stars":23,"repoUrl":24,"updatedAt":571},"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},[568,569,570],{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13},"2026-07-16T06:01:49.401142",{"slug":573,"name":573,"fn":574,"description":575,"org":576,"tags":577,"stars":23,"repoUrl":24,"updatedAt":582},"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},[578,579,580,581],{"name":499,"slug":500,"type":13},{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13},"2026-07-16T06:03:02.305907",{"slug":584,"name":584,"fn":585,"description":586,"org":587,"tags":588,"stars":23,"repoUrl":24,"updatedAt":594},"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},[589,590,591,593],{"name":484,"slug":485,"type":13},{"name":21,"slug":22,"type":13},{"name":592,"slug":38,"type":13},"Performance",{"name":9,"slug":8,"type":13},"2026-07-19T05:38:36.321999",{"slug":596,"name":596,"fn":597,"description":598,"org":599,"tags":600,"stars":23,"repoUrl":24,"updatedAt":604},"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},[601,602,603],{"name":21,"slug":22,"type":13},{"name":592,"slug":38,"type":13},{"name":9,"slug":8,"type":13},"2026-07-16T06:00:15.357243",{"slug":606,"name":606,"fn":607,"description":608,"org":609,"tags":610,"stars":23,"repoUrl":24,"updatedAt":615},"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},[611,612,613,614],{"name":484,"slug":485,"type":13},{"name":21,"slug":22,"type":13},{"name":592,"slug":38,"type":13},{"name":9,"slug":8,"type":13},"2026-07-16T05:59:17.325839",{"slug":617,"name":617,"fn":618,"description":619,"org":620,"tags":621,"stars":23,"repoUrl":24,"updatedAt":628},"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},[622,623,626,627],{"name":484,"slug":485,"type":13},{"name":624,"slug":625,"type":13},"Migration","migration",{"name":592,"slug":38,"type":13},{"name":9,"slug":8,"type":13},"2026-07-16T06:00:48.841055",30,{"items":631,"total":629},[632,639,646,653,660,667,674],{"slug":478,"name":478,"fn":479,"description":480,"org":633,"tags":634,"stars":23,"repoUrl":24,"updatedAt":491},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[635,636,637,638],{"name":484,"slug":485,"type":13},{"name":21,"slug":22,"type":13},{"name":488,"slug":489,"type":13},{"name":9,"slug":8,"type":13},{"slug":493,"name":493,"fn":494,"description":495,"org":640,"tags":641,"stars":23,"repoUrl":24,"updatedAt":506},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[642,643,644,645],{"name":499,"slug":500,"type":13},{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":504,"slug":505,"type":13},{"slug":508,"name":508,"fn":509,"description":510,"org":647,"tags":648,"stars":23,"repoUrl":24,"updatedAt":521},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[649,650,651,652],{"name":484,"slug":485,"type":13},{"name":515,"slug":516,"type":13},{"name":518,"slug":519,"type":13},{"name":9,"slug":8,"type":13},{"slug":523,"name":523,"fn":524,"description":525,"org":654,"tags":655,"stars":23,"repoUrl":24,"updatedAt":534},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[656,657,658,659],{"name":484,"slug":485,"type":13},{"name":530,"slug":531,"type":13},{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13},{"slug":536,"name":536,"fn":537,"description":538,"org":661,"tags":662,"stars":23,"repoUrl":24,"updatedAt":548},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[663,664,665,666],{"name":542,"slug":543,"type":13},{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":547,"slug":40,"type":13},{"slug":550,"name":550,"fn":551,"description":552,"org":668,"tags":669,"stars":23,"repoUrl":24,"updatedAt":561},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[670,671,672,673],{"name":556,"slug":557,"type":13},{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13},{"slug":563,"name":563,"fn":564,"description":565,"org":675,"tags":676,"stars":23,"repoUrl":24,"updatedAt":571},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[677,678,679],{"name":484,"slug":485,"type":13},{"name":9,"slug":8,"type":13},{"name":18,"slug":19,"type":13}]