id stringlengths 36 36 | url stringclasses 504
values | title stringclasses 475
values | section stringlengths 0 148 | content_type stringclasses 3
values | has_code bool 2
classes | text stringlengths 147 22.5k | vector listlengths 384 384 |
|---|---|---|---|---|---|---|---|
b22a06b0-8483-5043-91c4-5e764e6fbda7 | /course/beginners/module-0/qdrant-cloud/ | Qdrant Setup | Good Practices | course | false | Qdrant Setup > Good Practices
A few habits worth starting now:
- Keep your key out of your code. Use an environment variable or a secrets manager.
- Rotate your API keys now and then from the cluster **Access** tab.
- Use HTTPS only, and tighten access before you expose a cluster to the public internet. | [
0.04408007487654686,
-0.04569222405552864,
-0.028781559318304062,
-0.006269305478781462,
0.033230509608983994,
-0.061664458364248276,
0.018404575064778328,
0.020240867510437965,
0.0476626381278038,
-0.01204911433160305,
0.025552310049533844,
0.005823564250022173,
0.060678716748952866,
-0.0... |
b23b71e7-d1e7-5da5-a495-e32fac44f545 | /documentation/search-tuning/how-to-tune-hybrid-search/ | How to Tune Hybrid Search in Qdrant | Confirm Fusion Beats Either Prefetch | documentation | false | How to Tune Hybrid Search in Qdrant > Confirm Fusion Beats Either Prefetch
| Dataset | Dense Alone | Sparse Alone | Both, RRF (`k=2`) | Over the Better One | Second Prefetch Cost |
|---|---|---|---|---|---|
| SciFact | 0.6239 | 0.6886 | 0.7175 | +0.0289 | +0.73 ms |
| ArguAna | 0.4905 | 0.4224 | 0.5216 | +0.0311 | +1.... | [
0.032215625047683716,
-0.0012444565072655678,
-0.07521475851535797,
-0.038104139268398285,
0.04454459622502327,
-0.03935135900974274,
0.042527686804533005,
0.053346842527389526,
0.056641384959220886,
-0.06296160817146301,
0.02809077873826027,
0.020117679610848427,
0.017874963581562042,
-0.... |
b287ce3f-ba46-52eb-83e5-57efe1a34c46 | /documentation/quickstart/ | Local Quickstart | Next steps | documentation | false | Local Quickstart > Next steps
Now you know how Qdrant works. Getting started with [Qdrant Cloud](/documentation/cloud-quickstart/) is just as easy. [Create an account](https://qdrant.to/cloud) and use our SaaS completely free. We will take care of infrastructure maintenance and software updates.
To move onto some mor... | [
0.026013556867837906,
-0.052275702357292175,
-0.04816754534840584,
-0.024100782349705696,
0.0628514289855957,
-0.034779444336891174,
0.10728945583105087,
0.01810085028409958,
0.08343588560819626,
-0.03456127643585205,
0.03211717680096626,
0.02741238661110401,
0.009236415848135948,
-0.04497... |
b2959562-a357-5aec-8b00-f082c734d689 | /documentation/private-cloud/api-reference/ | API Reference | qdrant.io/v1 > Resource Types > ComponentStatus | documentation | false | API Reference > qdrant.io/v1 > Resource Types > ComponentStatus
_Appears in:_
- [QdrantCloudRegionStatus](#qdrantcloudregionstatus)
| Field | Description | Default | Validation |
| --- | --- | --- | --- |
| `name` _string_ | Name specifies the name of the component | | |
| `namespace` _string_ | Namespace specifies... | [
0.0244643185287714,
-0.050613537430763245,
-0.006640295963734388,
-0.06901274621486664,
0.06948860734701157,
-0.05592510104179382,
0.04934524744749069,
0.03950393199920654,
0.049867723137140274,
-0.04158982262015343,
0.00880097970366478,
-0.0006795738590881228,
0.05433971807360649,
-0.0558... |
b29e1d77-d8f3-5f55-98b4-68155ac55605 | /documentation/private-cloud/qdrant-cluster-management/ | Managing a Cluster | Upgrading the Qdrant version | documentation | false | Managing a Cluster > Upgrading the Qdrant version
To upgrade the Qdrant version of a database cluster, update the `version` field in the QdrantCluster spec. The Qdrant operator will automatically upgrade the cluster to the new version. The upgrade process is highly available on a multi-node cluster with replicated col... | [
0.02233511209487915,
-0.05774032324552536,
-0.04227777570486069,
-0.017471151426434517,
0.0611770860850811,
-0.06247319281101227,
0.051570747047662735,
0.02585342526435852,
0.02426338382065296,
-0.032258372753858566,
0.04562097787857056,
0.03502838313579559,
0.05431830510497093,
-0.0481979... |
b2b07c57-eeda-5e32-b5e1-2fdc9dc7b5c4 | /course/beginners/module-2/ingestion-pipeline/ | Ingestion Pipeline: End-to-End | course | true | Ingestion Pipeline: End-to-End
Module 2
Let's put everything together. This section walks through the complete ingestion pipeline from cloud setup to your first query.
### Step 1: Connect to Your Cluster
Module 0 walks you through creating a free cluster at [Qdrant Cloud](https://cloud.qdrant.io/) and retrieving it... | [
0.037627171725034714,
-0.03384198620915413,
-0.06430888175964355,
-0.02162051759660244,
0.06510508805513382,
-0.03748255968093872,
0.054682645946741104,
0.05440020188689232,
0.07410863041877747,
-0.03939511254429817,
0.022529518231749535,
0.02720065787434578,
0.030959252268075943,
-0.03336... | |
b2c4157b-6498-56ef-b0fb-2971dea1054d | /documentation/production-operations/managed-cloud-prometheus/ | Managed Cloud Prometheus Monitoring | Cluster metrics versus node metrics | documentation | true | Managed Cloud Prometheus Monitoring > Cluster metrics versus node metrics
Qdrant Managed Cloud exposes two Prometheus-compatible endpoints, and they answer different questions:
- `/metrics` reports on a single Qdrant node: collection counts, point and vector counts, API response times, and process-level memory such a... | [
0.03716844320297241,
-0.024343078956007957,
-0.04526481777429581,
-0.017511432990431786,
0.04101969301700592,
-0.027644982561469078,
0.07064066082239151,
0.05027694255113602,
0.08140105754137039,
-0.043938860297203064,
0.036678239703178406,
-0.004210493061691523,
0.057586900889873505,
-0.0... |
b2c69659-0fb0-54a7-aa06-bb9eecb71e3a | /documentation/tutorials-build-essentials/rag-deepseek/ | 5-Minute RAG with DeepSeek | Overview | documentation | false | 5-Minute RAG with DeepSeek > Overview
In this tutorial, we will:
1. Take sample text and turn it into vectors with FastEmbed.
2. Send the vectors to a Qdrant collection.
3. Connect Qdrant and DeepSeek into a minimal RAG pipeline.
4. Ask DeepSeek different questions and test answer accuracy.
5. Enrich DeepSeek prompts... | [
0.05360252410173416,
-0.060520317405462265,
-0.02934872917830944,
-0.04296274855732918,
0.07402290403842926,
-0.03424155339598656,
0.07359988987445831,
0.03269752115011215,
0.03496863320469856,
-0.05053291469812393,
0.0078039392828941345,
0.008584532886743546,
0.06622644513845444,
-0.01632... |
b2c9a128-7712-55ab-bf8a-c4fb49b73688 | /documentation/hybrid-cloud/platform-deployment-options/ | Deployment Platforms | Red Hat OpenShift | documentation | false | Deployment Platforms > Red Hat OpenShift
[Red Hat OpenShift Kubernetes Engine](https://www.redhat.com/en/technologies/cloud-computing/openshift/kubernetes-engine) provides you with the basic functionality of Red Hat OpenShift. It offers a subset of the features that Red Hat OpenShift Container Platform offers, like fu... | [
0.02314005047082901,
-0.04317222163081169,
-0.031578268855810165,
-0.027229901403188705,
0.06774770468473434,
-0.04495721310377121,
0.04632258042693138,
0.03913785517215729,
0.06359640508890152,
-0.0383186973631382,
0.04425801709294319,
-0.004704985301941633,
0.0157844889909029,
-0.0419510... |
b2cfc6b7-54bd-5d76-b197-07a05a5b56a6 | /documentation/tutorials-operations/secure-and-expose-hybrid-cloud/ | Securing and Exposing a Hybrid Cloud Cluster | documentation | false | Securing and Exposing a Hybrid Cloud Cluster
A new Qdrant cluster on Hybrid Cloud starts with no API key and no TLS, and that's by design: the cluster only lives on a `ClusterIP` Service inside your Kubernetes network by default, so it isn't reachable from outside your cluster in the first place. The moment you expose... | [
0.03948972746729851,
0.001945548108778894,
-0.0686652660369873,
-0.04554959759116173,
0.05401460453867912,
-0.05712495371699333,
0.07431989163160324,
0.01877060905098915,
0.04767046868801117,
-0.03419148921966553,
0.018145186826586723,
0.027724720537662506,
0.06522625684738159,
0.004196047... | |
b2ef9e2b-e1a4-5a49-980f-31b69e913bc2 | /course/beginners/module-6/inspecting-a-collection/ | Inspecting a Collection | course | false | Inspecting a Collection
Module 6
Two features for checking what you actually ingested.
| Feature | What it does |
|---------|--------------|
| [Facet counts](/documentation/manage-data/payload/#facet-counts) | Counts how many points hold each value of a payload field, which also shows how selective a filter would be... | [
0.06245848909020424,
-0.039731141179800034,
-0.0470350906252861,
-0.08455968648195267,
0.03343861922621727,
-0.01673000305891037,
0.03717989847064018,
0.05760488659143448,
0.07963047921657562,
-0.03292703628540039,
0.020688068121671677,
0.038221027702093124,
0.05606521666049957,
0.00012602... | |
b2f9a95d-5673-5c59-a212-f633bf931a72 | /documentation/migrate-to-qdrant/from-faiss/ | From FAISS | Supported Index Types | documentation | false | From FAISS > Supported Index Types
| FAISS Index Type | Supported | Notes |
| :--- | :--- | :--- |
| `IndexFlatL2` | Yes | Maps to `euclid` distance |
| `IndexFlatIP` | Yes | Maps to `dot` distance |
| `IndexHNSWFlat` | Yes | Full vectors are stored |
| `IndexIVFFlat` | Yes | Full vectors are stored |
| `IndexIVFPQ` |... | [
0.05938996747136116,
-0.02607358619570732,
-0.013213631696999073,
-0.036961376667022705,
0.030471768230199814,
-0.028116099536418915,
0.019409265369176865,
0.025547074154019356,
0.09760446101427078,
-0.043050020933151245,
0.04222085699439049,
0.013591358438134193,
0.040263786911964417,
-0.... |
b3188c38-fdba-56ec-a816-1e437e34fa29 | /course/essentials/ | Qdrant Essentials Course | Time commitment | course | false | Qdrant Essentials Course > Time commitment
- Duration: 6 days at 1-2 hours/day + 1 optional bonus day
- Video learning: ~3 hours
- Hands-on learning: 4-5 hours
- Final project: 2-4 hours
- Total: 9-12 hours
**What you’ll get**
- Build a production-ready docs search engine
- Practice with real projects
- Learn perform... | [
0.009199670515954494,
-0.04152475669980049,
-0.03926871716976166,
-0.010509451851248741,
0.04342713579535484,
-0.03357717767357826,
0.029862016439437866,
0.042481232434511185,
0.11811507493257523,
-0.04056517034769058,
0.02636418119072914,
0.001285972655750811,
0.06002349779009819,
-0.0750... |
b31a4c12-4d43-57ad-98b5-4aee436de6c2 | /articles/filterable-hnsw/ | Filterable HNSW Without Recall Loss | Conclusion | article | false | Filterable HNSW Without Recall Loss > Conclusion
It is possible to enchant HNSW algorithm so that it will support filtering points in a first search phase.
Filtering can be carried out on the basis of belonging to categories,
which in turn is generalized to such popular cases as numerical ranges and geo.
Experiments ... | [
0.054147761315107346,
-0.0332782082259655,
-0.018958767876029015,
-0.04863157123327255,
0.01852535642683506,
-0.037559475749731064,
0.07441652566194534,
0.05115272477269173,
0.06112360581755638,
0.015137559734284878,
0.04678265377879143,
0.06283731758594513,
0.10849250108003616,
0.00675143... |
b33abddf-46cb-5d81-bbd0-169e0e88d151 | /course/ | Qdrant Academy | Available Now | course | false | Qdrant Academy > Available Now
**What you'll gain:**
- Why Traditional Search Falls Short
- Embeddings and Distance Metrics
- Vector Search First Principles
- Sparse, Dense, and Hybrid Search
- Designing a Vector Search System
- Capstone: Multimodal Supplier Risk Intelligence
Time to Complete: under 5 hours
Includes:... | [
0.03775741159915924,
-0.029749158769845963,
-0.04096398130059242,
-0.012859202921390533,
0.06325129419565201,
-0.04540805518627167,
0.035029105842113495,
0.03991614654660225,
0.11323041468858719,
-0.016422266140580177,
0.03778596594929695,
0.030843134969472885,
0.011479856446385384,
-0.041... |
b34dd7c8-00f1-5f66-aa23-0dc903ec156f | /articles/sparse-vectors/ | Understanding SPLADE and Sparse Vectors | article | false | Understanding SPLADE and Sparse Vectors
Think of a library with a vast index card system. Each index card only has a few keywords marked out (sparse vector) of a large possible set for each book (document). This is what sparse vectors enable for text. | [
0.03243682160973549,
-0.03995699808001518,
-0.06053854525089264,
-0.04414959251880646,
0.047068040817976,
-0.04527222737669945,
0.070418581366539,
0.028771724551916122,
0.04987414553761482,
-0.014285964891314507,
0.014436392113566399,
0.03444879502058029,
0.07258985936641693,
0.01335292495... | |
b381c5f5-b219-515e-9509-c7250a99809b | /documentation/private-cloud/api-reference/ | API Reference | qdrant.io/v1 > Resource Types > TraefikConfig | documentation | false | API Reference > qdrant.io/v1 > Resource Types > TraefikConfig
_Appears in:_
- [Ingress](#ingress)
| Field | Description | Default | Validation |
| --- | --- | --- | --- |
| `allowedSourceRanges` _string array_ | AllowedSourceRanges specifies the allowed CIDR source ranges for the ingress. | | Optional: \{\} |
| `en... | [
0.05014701560139656,
-0.06889161467552185,
-0.021256599575281143,
-0.050032954663038254,
0.07790949195623398,
-0.03035777620971203,
0.051775939762592316,
0.08328201621770859,
0.010614086873829365,
-0.01867232844233513,
0.03179890289902687,
-0.0034885527566075325,
0.009883777238428593,
-0.0... |
b382f6e8-f519-5c1d-a6ff-8cd929e60544 | /documentation/edge/edge-api/reading-data/ | Reading Data | Request Builders | documentation | true | Reading Data > Request Builders
In Rust, the request types follow the fluent builder pattern:
```rust
let request = QueryRequestBuilder::new(10)
.with_payload(WithPayloadInterface::Bool(true))
.build();
```
Builders are available for `QueryRequest`, `ScrollRequest`, `RetrieveRequest`, `CountRequest`, `FacetR... | [
-0.0009930413216352463,
-0.024805648252367973,
-0.027116527780890465,
-0.030682217329740524,
0.03443264216184616,
-0.06576209515333176,
0.04999668896198273,
0.04698748141527176,
0.047647539526224136,
-0.03285367786884308,
0.06971950829029083,
0.044306084513664246,
0.054702144116163254,
-0.... |
b39a9926-38bf-540f-bc89-d11e4e07a7aa | /documentation/search-tuning/when-your-collection-outgrows-ram/ | When Your Collection Outgrows RAM | What to Tune Next | documentation | false | When Your Collection Outgrows RAM > What to Tune Next
Keep the first configuration that meets your held-out `nDCG@k` and latency targets. If none qualifies, test another quantization method or a higher `oversampling` value.
On a multi-shard hybrid collection, rerun the full request on your deployed shard layout once ... | [
0.02584538795053959,
-0.0496409609913826,
-0.03532952070236206,
-0.04942500591278076,
0.06512662023305893,
-0.04417603462934494,
0.07213302701711655,
0.06466439366340637,
0.0915379747748375,
-0.028446139767766,
0.013632571324706078,
0.027700265869498253,
0.052807558327913284,
-0.0217151157... |
b39f561a-f959-599d-a968-6c380b9bc0de | /documentation/tutorials-search-engineering/query-decomposition/ | Query Decomposition for Multi-Hop Questions | How It Works | documentation | true | Query Decomposition for Multi-Hop Questions > How It Works
- **Interleaving**: take each hop's first chunk, then each hop's second, and so on, skipping duplicates, until you fill up a limit of chunks you'll show to the LLM.
- **Highest score**: keep each chunk once with its best similarity score and sort. It's simple,... | [
0.0379607118666172,
-0.040960103273391724,
-0.051753778010606766,
-0.03486373648047447,
0.07400619238615036,
-0.017425507307052612,
0.030212584882974625,
0.010554243810474873,
0.06249554082751274,
-0.013876673765480518,
0.04488113150000572,
-0.0167628675699234,
0.05594252794981003,
-0.0129... |
b3a665a1-1e2a-5bf8-a9c7-28dbc1801177 | /documentation/search/explore/ | Explore | Recommendation API > Sum Scores Strategy | documentation | false | Explore > Recommendation API > Sum Scores Strategy
Another strategy for using multiple query vectors simultaneously is to just sum their scores against the candidates. In qdrant, this is called `sum_scores` strategy.
This strategy was used in [this paper](https://arxiv.org/abs/2210.10695) by [UKP Lab](http://www.ukp.... | [
0.04878051578998566,
-0.04540516436100006,
-0.025897720828652382,
-0.0125211076810956,
0.03185886889696121,
-0.017649294808506966,
0.06337375938892365,
0.01490149088203907,
0.07466874271631241,
-0.02515658177435398,
0.03382673114538193,
0.02900332398712635,
0.08378677070140839,
-0.01202898... |
b3a93eb8-a985-5412-b7a1-29158970731f | /documentation/embeddings/nomic/ | Nomic | Run Locally with FastEmbed | documentation | true | Nomic > Run Locally with FastEmbed
[FastEmbed](/documentation/fastembed/) runs `nomic-ai/nomic-embed-text-v1.5` and `nomic-ai/nomic-embed-vision-v1.5` on your CPU with ONNX Runtime. The models download from Hugging Face on first use, and no API key is needed. `nomic-ai/nomic-embed-text-v1.5-Q` is a quantized variant a... | [
0.05969960242509842,
-0.05715009197592735,
-0.05393627658486366,
-0.03076126053929329,
0.08006235212087631,
-0.04318122938275337,
0.06019234657287598,
0.014495466835796833,
0.01577380672097206,
-0.01714986003935337,
0.042370185256004333,
0.04579796642065048,
0.044850241392850876,
-0.056114... |
b3aa5b65-fbe3-5656-b494-2168fec00bbe | /documentation/inference/ | Inference | Inference Options | documentation | false | Inference > Inference Options
Where and how you generate embeddings depends on the model you want to use and whether you prefer to manage your own embedding infrastructure:
- **Client-side inference**: Manage your own inference pipeline locally, for example, using Qdrant's Python [FastEmbed](/documentation/fastembed/... | [
0.03016320802271366,
-0.019429882988333702,
-0.032085251063108444,
-0.06685923784971237,
0.05585980787873268,
-0.038393694907426834,
0.07310181856155396,
0.04009599983692169,
0.0493665486574173,
-0.035749536007642746,
0.04986985772848129,
-0.0010559939546510577,
0.04496144503355026,
-0.041... |
b3bcb940-9b7b-5a21-9f4a-282836fc1ec7 | /documentation/fastembed/fastembed-splade/ | Working with SPLADE | Examine weights | documentation | true | Working with SPLADE > Examine weights
Now, print the first 5 features and their weights for better understanding.
```python
for i in range(5):
print(f"Token at index {sparse_embeddings_list[0].indices[i]} has weight {sparse_embeddings_list[0].values[i]}")
```
The output will display the token indices and their c... | [
0.029270818457007408,
-0.050231169909238815,
-0.01816818304359913,
-0.08696803450584412,
0.05657271668314934,
-0.03409635275602341,
-0.008159197866916656,
0.0636800155043602,
0.06901600956916809,
-0.027030138298869133,
0.01663517951965332,
0.03707874193787575,
0.05745068937540054,
-0.00248... |
b3c2139a-e3d2-5e89-b256-a99d3db5388e | /documentation/tutorials-develop/serverless-semantic-search/ | Serverless Semantic Search | Create the Project | documentation | true | Serverless Semantic Search > Create the Project
Create a Rust library project, and replace the generated `Cargo.toml` with the following. Cargo builds every file in `src/bin` as a binary. You will add two: `setup` stores your documents, and `search` is the Lambda function.
```bash
cargo new semantic-search --lib
cd s... | [
0.03680777549743652,
-0.059563349932432175,
-0.050716206431388855,
-0.021769138053059578,
0.08145829290151596,
-0.04470330476760864,
0.02082609012722969,
0.005535745061933994,
0.018555928021669388,
-0.03681117668747902,
0.04617549106478691,
-0.011545921675860882,
0.03677530586719513,
-0.01... |
b3e11290-c8b1-59b6-b8ee-0589bd4ecc68 | /course/multi-vector-search/module-2/colpali-family/ | ColPali Family Overview | Benchmarking Visual Document Retrieval | course | false | ColPali Family Overview > Benchmarking Visual Document Retrieval
The **ViDoRe (Visual Document Retrieval) Benchmark** has emerged as the leading evaluation framework for visual retrieval models on document understanding tasks. As of January 2026, it stands as the largest and most comprehensive benchmark in the field, ... | [
0.03010367415845394,
-0.007871950976550579,
-0.03813706710934639,
-0.04463572055101395,
0.09758009016513824,
-0.042234018445014954,
0.056525927037000656,
0.058845289051532745,
0.0670926496386528,
-0.03690663352608681,
0.01536471489816904,
-0.015045233070850372,
0.021906131878495216,
-0.040... |
b402698c-a33e-55c5-a8ab-4352b20cf3c4 | /course/multi-vector-search/module-3/quantization-techniques/ | Vector Quantization Techniques | Enabling Quantization in Qdrant | course | true | Vector Quantization Techniques > Enabling Quantization in Qdrant
One of Qdrant's powerful features: **you can enable quantization on an existing collection** without changing your inference or ingestion pipelines. The quantization happens transparently during indexing.
First, let's load the ColPali model and prepare ... | [
0.060644906014204025,
-0.048750657588243484,
-0.03090902417898178,
-0.062096159905195236,
0.03863486275076866,
-0.02251567132771015,
0.05997548624873161,
0.05431888997554779,
0.09398507326841354,
-0.025313422083854675,
0.005743186455219984,
0.02992221713066101,
0.0634041354060173,
-0.01456... |
b4124e4b-c95a-5394-a686-43d52a9a81ea | /articles/what-is-a-vector-database/ | What is a Vector Database? | Distributed Deployment > Sharding: Distributing Data Across Nodes | article | true | What is a Vector Database? > Distributed Deployment > Sharding: Distributing Data Across Nodes
In a distributed Qdrant cluster, data is split into smaller units called **shards**, which are distributed across different nodes. This helps balance the load and ensures that queries can be processed in parallel.
Each coll... | [
0.019062891602516174,
-0.058607835322618484,
-0.06882311403751373,
-0.021948782727122307,
0.05035970360040665,
-0.03902457281947136,
0.06940263509750366,
0.034350745379924774,
0.0699177235364914,
-0.021493034437298775,
0.059021785855293274,
0.016565905883908272,
0.030659770593047142,
-0.00... |
b413d5b2-f6b9-5b27-a09a-c8f078615a03 | /course/multi-vector-search/module-3/final-project/ | Final Project: Build Your Own Multi-Vector Search System | What You've Accomplished | course | false | Final Project: Build Your Own Multi-Vector Search System > What You've Accomplished
By completing this project, you've built a production-ready multi-vector search system that:
* **Leverages late interaction** for superior search quality through token-level matching
* **Scales efficiently** using quantization, poolin... | [
0.020926926285028458,
-0.06202958524227142,
-0.037614598870277405,
-0.05723393335938454,
0.08404279500246048,
-0.03699088469147682,
0.05649237334728241,
0.03328212723135948,
0.07277388870716095,
-0.04214257374405861,
0.010634137317538261,
0.03720322996377945,
0.05414104089140892,
-0.047014... |
b41f0650-f245-567e-b364-a4b810ac1cd5 | /documentation/manage-data/points/ | Points | Scroll Points > Order Points by Payload Key | documentation | true | Points > Scroll Points > Order Points by Payload Key
_Available as of v1.8.0_
When using the [`scroll`](#scroll-points) API, you can sort the results by payload key. For example, you can retrieve points in chronological order if your payloads have a `"timestamp"` field, as shown in the following example:
Without an ... | [
0.027797868475317955,
-0.03808572515845299,
-0.025027217343449593,
-0.00906137004494667,
0.06366735696792603,
-0.05968017876148224,
0.06926681101322174,
0.04247720539569855,
0.09375257045030594,
-0.04860283061861992,
0.06146789714694023,
0.045380208641290665,
0.020693054422736168,
0.002006... |
b4346838-f2cd-5d84-aa25-3aa4a9395d4d | /documentation/tutorials-search-engineering/collaborative-filtering/ | Collaborative Filtering | documentation | false | Collaborative Filtering
| Time: 45 min | Level: Intermediate | [](https://githubtocolab.com/qdrant/examples/blob/master/collaborative-filtering/collaborative-filtering.ipynb) | |
|--------------|---------------------|--|----|
Every time Spot... | [
0.03977356478571892,
0.015381217002868652,
-0.06305838376283646,
-0.05205417424440384,
0.08484864234924316,
-0.008831480517983437,
0.06363572180271149,
0.05812735855579376,
0.07281718403100967,
-0.005244333762675524,
0.03750872611999512,
-0.002036147750914097,
0.06723204255104065,
-0.04662... | |
b46388bd-e34b-519d-acae-00ac0d574f4d | /documentation/tutorials-build-essentials/agentic-rag-langgraph/ | Agentic RAG with LangGraph | Implementation > Brave Search | documentation | true | Agentic RAG with LangGraph > Implementation > Brave Search
To enhance search capabilities, we’ll integrate Brave Search. Visit the [Brave API](https://api.search.brave.com/) and complete their API access request process to obtain an API key. This key will enable web search functionality for our agent.
For added secur... | [
0.0061089531518518925,
-0.04988739639520645,
-0.04822955280542374,
-0.0399780236184597,
0.0710766464471817,
-0.06788540631532669,
0.04996766895055771,
0.021549301221966743,
0.02084936946630478,
-0.023407280445098877,
0.016132071614265442,
0.05803706869482994,
0.07813018560409546,
-0.010437... |
b48ca5e7-28cd-581e-a721-bf09b7f22c79 | /articles/turboquant-quantization/ | TurboQuant in Qdrant | Further Reading | article | false | TurboQuant in Qdrant > Further Reading
**Engineering deep-dives by the team that built this:**
* **[TurboQuant + RaBitQ: a hybrid approach in Qdrant](https://ivanpleshkov.dev/blog/turboquant/)**, by [Ivan Pleshkov](https://www.linkedin.com/in/ivan-pleshkov/). Every algorithmic extension explained: renormalization, th... | [
0.05403142049908638,
-0.009048555046319962,
-0.07268142700195312,
-0.04630745202302933,
0.05041433125734329,
0.0067200190387666225,
0.05407007411122322,
0.014220351353287697,
0.09180499613285065,
-0.0492512546479702,
0.04945705831050873,
0.006762499455362558,
0.04194005951285362,
-0.045097... |
b498b383-cd21-5723-8265-e84dec8e1735 | /documentation/tutorials-basics/multimodal-search/ | Multimodal Search | Setup | documentation | true | Multimodal Search > Setup
Install the client:
```python
pip install qdrant-client
```
You need a Cohere API key to follow along. Create a free one on the Cohere dashboard. | [
0.022677920758724213,
-0.04681658372282982,
-0.019948294386267662,
-0.012558466754853725,
0.05345424264669418,
-0.04877304285764694,
0.014647766947746277,
0.004954909905791283,
0.016002120450139046,
-0.02694294974207878,
0.01799745112657547,
0.007015030365437269,
0.09726560115814209,
-0.04... |
b49e3327-fd89-518a-9272-821dd414ad43 | /course/essentials/day-1/embedding-models/ | Points, Vectors and Payloads | course | false | Points, Vectors and Payloads
Day 1
Understanding Qdrant's core data model is essential for building effective vector search applications. This lesson establishes the precise technical vocabulary and concepts you'll use throughout the course. | [
0.03769104182720184,
-0.045931220054626465,
-0.03231402114033699,
-0.03950141370296478,
0.021315941587090492,
-0.050057847052812576,
0.08600419014692307,
0.03187413886189461,
0.0885968878865242,
-0.008549508638679981,
0.0023358555044978857,
0.018843581900000572,
0.05152459442615509,
-0.051... | |
b4a1aef7-bbd1-5566-b192-39f3b7b98865 | /articles/minicoil/ | miniCOIL: on the Road to Usable Sparse Neural Retrieval | The Good, the Bad and the Ugly > Term-based Retrieval | article | false | miniCOIL: on the Road to Usable Sparse Neural Retrieval > The Good, the Bad and the Ugly > Term-based Retrieval
Term-based retrieval usually treats text as a bag of words. These words play roles of different importance, contributing to the overall relevance score between a document and a query.
Famous **BM25** estima... | [
0.003380111651495099,
-0.033885370939970016,
-0.044327035546302795,
-0.05742980167269707,
0.04140938073396683,
-0.0030422527343034744,
0.043857794255018234,
0.0371980220079422,
0.08768506348133087,
0.0009698711219243705,
0.03969269618391991,
-0.0019280601991340518,
0.07333441078662872,
-0.... |
b4ad756f-5b59-5ec0-8889-9cb9371113d3 | /documentation/search-tuning/before-tuning-a-qdrant-collection/ | What to Check Before Tuning a Qdrant Collection | Check the Winner on Fresh Queries | documentation | false | What to Check Before Tuning a Qdrant Collection > Check the Winner on Fresh Queries
A setting selected and evaluated on the same queries will look better than it performs on fresh queries. Split the labeled queries in half: select the winner on one half, then measure its gain on the other. We repeated that split 200 t... | [
0.02676057070493698,
-0.05530763417482376,
-0.060377079993486404,
0.010324081405997276,
0.04458237811923027,
-0.033338189125061035,
0.0472821481525898,
0.034957155585289,
0.0280818659812212,
-0.019792338833212852,
0.024911588057875633,
0.04085821285843849,
0.031821176409721375,
-0.03001153... |
b4bebbaa-cbf1-5ad6-ab70-c4d31171180c | /course/essentials/day-2/ | Day 2: Indexing and Performance | course | false | Day 2: Indexing and Performance
Day 2
Master [HNSW](/articles/filterable-hnsw/) indexing and practical tuning for fast retrieval.
---
## Today’s path
1. HNSW Indexing Fundamentals
2. Combining Vector Search and Filtering
3. Demo: HNSW Performance Tuning
4. Project: HNSW Performance Benchmarking
You’ll measure rec... | [
0.03315991535782814,
-0.023030703887343407,
-0.039625413715839386,
-0.02659996971487999,
0.0269259475171566,
-0.031576741486787796,
0.03274485096335411,
0.02885538525879383,
0.07748053222894669,
-0.01824936829507351,
0.01860189437866211,
0.07753419876098633,
0.06094690039753914,
-0.0393223... | |
b4d1cf3f-d1d0-50cf-a20b-df9e6ba26fc6 | /documentation/manage-data/quantization/ | Quantization | documentation | false | Quantization
Quantization is an optional feature in Qdrant that enables efficient storage and search of high-dimensional vectors.
By transforming original vectors into new representations, quantization compresses data while preserving close to original relative distances between vectors.
Different quantization methods... | [
0.038081370294094086,
-0.051182445138692856,
-0.06412094086408615,
-0.03469408303499222,
0.042638812214136124,
-0.003995146136730909,
0.06809039413928986,
0.019986450672149658,
0.12120860815048218,
-0.02616414614021778,
0.027234528213739395,
-0.00439428212121129,
0.05078738182783127,
-0.01... | |
b4d24205-dda7-5b7c-b2a7-3989cfbbe7ab | /documentation/frameworks/lakechain/ | AWS Lakechain | documentation | true | AWS Lakechain
[Project Lakechain](https://awslabs.github.io/project-lakechain/) is a framework based on the AWS Cloud Development Kit (CDK), allowing to express and deploy scalable document processing pipelines on AWS using infrastructure-as-code. It emphasizes on modularity and extensibility of pipelines, and provide... | [
0.017209317535161972,
-0.020018137991428375,
-0.055452074855566025,
-0.03939257934689522,
0.03287428244948387,
-0.02963564544916153,
0.053787555545568466,
0.01225600577890873,
0.03676373139023781,
-0.027538757771253586,
-0.004782939329743385,
-0.00782888662070036,
0.04376975819468498,
-0.0... | |
b4dc5469-5223-5f55-a693-cf2ec41fd455 | /articles/vector-similarity-beyond-search/ | Vector Similarity: Going Beyond Full-Text Search | Recommendations > Feature-Based Recommendations | article | true | Vector Similarity: Going Beyond Full-Text Search > Recommendations > Feature-Based Recommendations
The first approach is to take all positive and negative examples and average them to create a single query vector.
Qdrant averages the positive and the negative examples separately and combines the two averages into one ... | [
0.03635525703430176,
-0.06639403104782104,
-0.038108792155981064,
-0.034788329154253006,
0.04903404042124748,
-0.017064984887838364,
0.07885377109050751,
0.0433279387652874,
0.07758530974388123,
-0.03138153627514839,
0.02136705443263054,
0.01566123403608799,
0.03921893984079361,
-0.0185548... |
b500c4c6-0447-5808-9669-5ca667d91912 | /documentation/ops-configuration/quotas/ | Resource Quotas | documentation | false | Resource Quotas
*Available as of v1.19.0*
Resource quotas stop a node from taking on more data once it's low on memory or disk space. When usage reaches a configured limit, the node rejects updates that would consume more of that resource until usage drops back down. | [
0.0037339332047849894,
-0.023134570568799973,
-0.005011661909520626,
-0.07986634224653244,
0.052761733531951904,
-0.03507550433278084,
0.02845592610538006,
0.022369837388396263,
0.103756844997406,
-0.027987875044345856,
0.01911058835685253,
-0.0040233018808066845,
0.061453938484191895,
-0.... | |
b50d2031-5908-5902-a5a8-b80a17ac0d78 | /documentation/datasets/ | Practice Datasets | documentation | false | Practice Datasets
You may find that creating embeddings from datasets is a very resource-intensive task.
If you need a practice dataset, feel free to pick one of the ready-made snapshots on this page.
These snapshots contain pre-computed vectors that you can easily import into your Qdrant instance. | [
0.03836928308010101,
-0.027757398784160614,
-0.04125288501381874,
-0.05923619493842125,
0.039204929023981094,
-0.049292586743831635,
0.049729857593774796,
0.0555114708840847,
0.08318859338760376,
-0.04388153925538063,
0.022570934146642685,
0.03549255430698395,
0.06143202260136604,
-0.04211... | |
b511be3c-0efb-58dd-97f2-8acd314d0dc4 | /course/multi-vector-search/module-3/multi-stage-retrieval/ | Multi-Stage Retrieval with Universal Query API | course | false | Multi-Stage Retrieval with Universal Query API
Module 3
The most effective production deployments combine multiple optimization techniques in multi-stage pipelines. Fast approximate methods retrieve candidates, which are then reranked with higher-quality methods.
Qdrant's Universal Query API makes it easy to build s... | [
0.010871917940676212,
-0.041557371616363525,
-0.010223283432424068,
-0.025963548570871353,
0.05912042409181595,
-0.040720660239458084,
0.069814532995224,
0.03599904477596283,
0.06424616277217865,
-0.014742718078196049,
0.03672335669398308,
0.028465524315834045,
0.07976929098367691,
-0.0482... | |
b52ba345-510f-5f6d-ad57-bbea792bcc9b | /course/essentials/day-7/camel/ | Integrating with Camel AI | What You'll Learn | course | false | Integrating with Camel AI > What You'll Learn
Day 7
Agentic RAG with multi-agent systems using Camel AI and Qdrant.
- Multi-agent system architectures
- Agentic RAG patterns and best practices
- Agent collaboration and communication
- Building autonomous AI systems with Qdrant
- Auto-Retrieval with CAMEL for automat... | [
0.03462924063205719,
-0.024369170889258385,
-0.05532347783446312,
-0.00799392070621252,
0.10251697152853012,
-0.0494171604514122,
0.04489557072520256,
0.0178845152258873,
0.02705170214176178,
-0.013876525685191154,
0.05351821333169937,
-0.017071275040507317,
0.04852493852376938,
-0.0424568... |
b53c890c-1d47-55c2-b7fd-17e98b685598 | /documentation/search/filtering/ | Filtering | Filtering Conditions > Nested Key | documentation | true | Filtering > Filtering Conditions > Nested Key
*Available as of v1.1.0*
Payloads are JSON objects. A field you want to filter on can be nested inside another object or an array. To filter on a nested field, set the condition's `key` to the path of the field. The path syntax is similar to [jq](https://jqlang.github.io/... | [
0.04834327846765518,
-0.017709489911794662,
-0.04070620611310005,
-0.041092194616794586,
0.052702780812978745,
-0.01057297270745039,
0.025471156463027,
0.024738090112805367,
0.07675346732139587,
-0.03743091598153114,
0.04837992042303085,
0.033115535974502563,
0.04578300938010216,
0.0304140... |
b54012ea-ec60-5a9e-ba89-dfc7eaa07b0f | /documentation/tutorials-build-essentials/agentic-rag-crewai-zoom/ | Agentic RAG with CrewAI | Conclusion | documentation | false | Agentic RAG with CrewAI > Conclusion
This tutorial has demonstrated how to build a sophisticated meeting analysis system that combines vector search with AI agents. Let's recap the key components we've covered:
1. **Vector Search Integration**
- Efficient storage and retrieval of meeting content using Qdrant
- ... | [
0.00644278759136796,
-0.04205285757780075,
-0.07023705542087555,
-0.03123646415770054,
0.09769200533628464,
-0.047507911920547485,
0.05375449359416962,
0.031721632927656174,
0.0695275217294693,
-0.04381532222032547,
0.0019374563125893474,
-0.01846001110970974,
0.05047281086444855,
-0.04938... |
b56dfc37-08c3-597c-9155-fe7350a1d403 | /documentation/data-synchronization/with-postgres/ | Keeping Postgres and Qdrant in Sync | Tier 3: Change Data Capture with Debezium + Redpanda > Postgres Setup | documentation | true | Keeping Postgres and Qdrant in Sync > Tier 3: Change Data Capture with Debezium + Redpanda > Postgres Setup
```sql
-- Enable in postgresql.conf (or docker-compose command args):
-- wal_level = logical
-- Create a publication for the products table
CREATE PUBLICATION products_publication FOR TABLE products;
``` | [
0.04630528390407562,
-0.03408253192901611,
0.0030482325237244368,
-0.032370757311582565,
0.014806308783590794,
-0.05120593681931496,
0.05128166079521179,
0.03347994387149811,
0.0716758742928505,
-0.02124803513288498,
0.04524499177932739,
0.03055727481842041,
0.07857522368431091,
-0.0378952... |
b573b0cd-9a02-5604-8797-06f05e68aab8 | /documentation/edge/edge-api/updating-data/ | Updating Data | Update Operations | documentation | false | Updating Data > Update Operations
| Operation | Parameters | Description |
| --- | --- | --- |
| `clear_payload` | `point_ids` | Delete all payload from points. |
| `clear_payload_by_filter` | `filter` | Delete all payload from matching points. | | [
0.05050652474164963,
-0.026137420907616615,
-0.03228344768285751,
-0.031011080369353294,
0.06375927478075027,
0.004550570156425238,
0.008625919930636883,
0.05644424632191658,
0.06322017312049866,
-0.025207696482539177,
0.05385281518101692,
0.05277561768889427,
0.07008622586727142,
-0.00382... |
b58276d1-374c-54a4-b56f-6a9acc0398b0 | /documentation/search/text-search/text-filtering/ | Text Filtering | Filtering on an Exact String | documentation | true | Text Filtering > Filtering on an Exact String
To filter on exact strings, first create a [payload index](/documentation/manage-data/indexing/#payload-index) of type `keyword`for the field you want to filter on. A payload index makes filtering faster and reduces the load on the system.
Filtering on a field without an ... | [
0.011910990811884403,
-0.034098900854587555,
-0.04189855605363846,
0.0067116557620465755,
0.0705786645412445,
-0.011018626391887665,
0.04448345676064491,
0.03289690241217613,
0.07168884575366974,
-0.033457014709711075,
0.024805957451462746,
0.03991163522005081,
0.058562517166137695,
-0.019... |
b583eef1-9e57-501b-ab4e-edf74fa8b40f | /documentation/tutorials-search-engineering/sparse-embeddings-ecommerce-part-2/ | Fine-Tuning Sparse Embeddings for E-Commerce Search | Part 2: Training SPLADE on Modal | The Dataset: Amazon ESCI | documentation | true | Fine-Tuning Sparse Embeddings for E-Commerce Search | Part 2: Training SPLADE on Modal > The Dataset: Amazon ESCI
We use Amazon's [ESCI dataset](https://github.com/amazon-science/esci-data) (Shopping Queries Dataset), released for KDD Cup 2022. It's one of the most realistic e-commerce search benchmarks available:
- ... | [
0.047642022371292114,
-0.006033670622855425,
-0.07296161353588104,
-0.05392652004957199,
0.03999301418662071,
-0.05137960985302925,
0.05131989344954491,
0.02555542066693306,
0.07133133709430695,
-0.03402801603078842,
0.01582368090748787,
0.018292004242539406,
0.06361057609319687,
-0.016083... |
b58e1879-6d6a-5d0c-b4b4-b8ccf3cc35bb | /documentation/hybrid-cloud/configure-scale-upgrade/ | Configure, Scale & Update Clusters | documentation | false | Configure, Scale & Update Clusters
## Configure Clusters
Alongside Hybrid Cloud specific scheduling options, you can also adjust various other advanced configuration options for your clusters. See [Configure Clusters](/documentation/cloud/configure-cluster/) for more details.
## Scale Clusters
Hybrid cloud clusters... | [
0.034253381192684174,
0.000715811678674072,
-0.05563798546791077,
-0.04636388644576073,
0.08148675411939621,
-0.0553828664124012,
0.061017561703920364,
0.06922543048858643,
0.07130037248134613,
-0.04811687022447586,
0.0281346645206213,
-0.0036701734643429518,
0.024309132248163223,
-0.02947... | |
b5b4148a-549c-53e5-b0ed-cb75de92ab2e | /documentation/cloud/configure-cluster/ | Configure Clusters | Rename a Cluster | documentation | false | Configure Clusters > Rename a Cluster
You can rename a Qdrant Cluster from the cluster's detail page.

Renaming a cluster does not affect its functionality or configuration. The cluster's unique ID and cluster URLs will remain the same.
 on all the fields you're filtering by. Payload indexing often produces larger speedups for filtered queries than other optimizations such as changes to ... | [
0.03686768561601639,
-0.021407058462500572,
-0.06140727922320366,
-0.025929691269993782,
0.05347081646323204,
-0.03341711312532425,
0.03459162265062332,
0.021546242758631706,
0.08854778856039047,
-0.04571428894996643,
0.02086729370057583,
0.023532869294285774,
0.05092671141028404,
-0.02985... |
b5c8e10b-60a4-5546-afc3-b6e1cd7792e8 | /documentation/embeddings/cohere/ | Cohere | documentation | false | Cohere
[Cohere](https://docs.cohere.com/docs/cohere-embed) provides multimodal embedding models through its [Embed API](https://docs.cohere.com/reference/embed). There are two ways to use them with Qdrant:
- Call the Cohere API from your application and send the vectors to Qdrant. This works with any Qdrant deploymen... | [
0.05809316411614418,
-0.02985391952097416,
-0.006416474934667349,
-0.057085298001766205,
0.05690256133675575,
-0.026221346110105515,
0.09352705627679825,
0.058929380029439926,
0.037020981311798096,
-0.009565410204231739,
0.023189786821603775,
-0.014260176569223404,
0.04463060200214386,
-0.... | |
b5d82172-c052-5973-97c7-3480a79b2505 | /documentation/tutorials-operations/incremental-embedding-updates/ | Incremental Embedding Updates | Populate Collection | documentation | true | Incremental Embedding Updates > Populate Collection
Populate the collection with the whole documentation.
```python
client.upsert(COLLECTION, points=[
models.PointStruct(
id=c["point_id"],
vector=models.Document(text=c["text"], model=MODEL),
payload=payload(c),
)
for c in prepare_c... | [
0.06277846544981003,
-0.028641272336244583,
-0.007730032783001661,
-0.06512967497110367,
0.10537119954824448,
-0.03814699500799179,
0.012764379382133484,
0.02278037555515766,
0.05242704227566719,
-0.018411410972476006,
0.012576071545481682,
0.03343595564365387,
0.06772581487894058,
-0.0371... |
b5ddb35d-8366-5c03-98cc-66b14fdd356a | /articles/immutable-data-structures/ | Immutable Data Structures in Qdrant | How Does Immutability Help? > Defragmentation | article | false | Immutable Data Structures in Qdrant > How Does Immutability Help? > Defragmentation
When you read data from a disk, you almost never read a single byte. Instead, you read a page, which is a fixed-size chunk of data.
On many systems, the page size is 4KB, which means that every read operation will read 4KB of data, eve... | [
0.06346866488456726,
-0.05012943968176842,
-0.08327867090702057,
-0.038923125714063644,
0.06908898800611496,
-0.036115214228630066,
0.0496431402862072,
0.02250790223479271,
0.08769587427377701,
-0.010108733549714088,
-0.009381229057908058,
-0.007858268916606903,
0.04894551262259483,
-0.027... |
b5fbf6a5-c1d8-5236-babf-235ccbaccf20 | /articles/agentic-rag/ | What is Agentic RAG? Building Agents with Qdrant | Agentic RAG: Combining RAG with Agents > Solving Information Retrieval Problems with LLMs | article | false | What is Agentic RAG? Building Agents with Qdrant > Agentic RAG: Combining RAG with Agents > Solving Information Retrieval Problems with LLMs
Each of these tools might be a separate agent on its own, and multi-agent systems are not uncommon. In such cases,
agents can communicate with each other, and one agent can decid... | [
-0.008711266331374645,
-0.04338742420077324,
-0.08526724576950073,
-0.0227922685444355,
0.10356099903583527,
-0.05360369384288788,
0.07453510165214539,
0.038483474403619766,
0.05991039425134659,
-0.020281540229916573,
0.056368596851825714,
0.025614209473133087,
0.06049475446343422,
-0.0189... |
b61df486-d2c9-5623-a316-49ff50a9514b | /documentation/production-operations/bulk-data-import/ | Bulk Uploading Data to Qdrant | Confirm the Upload Has Finished | documentation | false | Bulk Uploading Data to Qdrant > Confirm the Upload Has Finished
The upload call returns before the optimizer finishes indexing. Before you serve production traffic, check that Qdrant has caught up:
- **Collection status:** `yellow` means the optimizer is still working, and `green` means it is done. A `grey` status me... | [
0.01763210818171501,
-0.032919786870479584,
-0.035099729895591736,
-0.040268201380968094,
0.04780353605747223,
-0.03480451926589012,
0.07998771220445633,
0.026980189606547356,
0.10094266384840012,
-0.046463895589113235,
0.018628422170877457,
0.03853563591837883,
0.020556509494781494,
-0.05... |
b61e59ff-80b1-55c5-99d1-4b980d06b520 | /documentation/embeddings/mistral/ | Mistral | Embed Documents | documentation | true | Mistral > Embed Documents
One request embeds a list of inputs. Each item in the response carries the `index` of its input, which this example uses as the point ID. The SDKs type `index` and `embedding` as optional, so the example asserts that they're set:
```python
documents = [
"Qdrant is a vector search engine.... | [
0.03996358811855316,
-0.03856228291988373,
-0.019482648000121117,
0.0010469384724274278,
0.04548760503530502,
-0.03458549082279205,
0.10595282167196274,
0.09054435044527054,
0.05106761306524277,
-0.013954509980976582,
0.020028777420520782,
0.05664881691336632,
0.03451770916581154,
-0.03663... |
b6263b3e-7f4c-5558-b99c-b24e6066108d | /course/beginners/module-0/building-simple-vector-search/ | Implementing a Basic Vector Search | Step 4: Create a Collection | course | true | Implementing a Basic Vector Search > Step 4: Create a Collection
A [collection](/documentation/manage-data/collections/) is where your vectors live. It's a lot like a table in a regular database: a named container for related data. When you create one, you tell Qdrant two things:
- **Size:** how many numbers each vec... | [
0.02766905166208744,
-0.04454497992992401,
-0.03049299865961075,
-0.010787641629576683,
0.07604861259460449,
-0.028628524392843246,
0.06053202971816063,
0.06761225312948227,
0.10858985781669617,
-0.0529942661523819,
0.03885982185602188,
0.02501595951616764,
0.03322330117225647,
-0.02260231... |
b628bd73-057e-51f8-9e1a-0eaed5436972 | /documentation/embeddings/mixedbread/ | Mixedbread | Setup | documentation | true | Mixedbread > Setup
Install sentence-transformers and the Qdrant client:
```bash
pip install sentence-transformers qdrant-client
```
Load the model and create a collection sized for it. The first run downloads the model from Hugging Face, about 670 MB for `mxbai-embed-large-v1`.
```python
from qdrant_client import Q... | [
0.02438029833137989,
-0.043947502970695496,
-0.02717623859643936,
-0.06999944150447845,
0.0672018751502037,
-0.04534177854657173,
0.03602705895900726,
0.036490876227617264,
0.045414675027132034,
-0.06422062963247299,
0.027294358238577843,
0.010347128845751286,
0.049629271030426025,
-0.0572... |
b65cd8cb-9975-5ea5-a259-b50578b66615 | /documentation/cloud-rbac/user-management/ | User Management | Removing a User from an Account | documentation | false | User Management > Removing a User from an Account
Users can be removed from an account by clicking on their name in either **User Management** (via Actions). This option is only available after they've accepted the invitation to join, ensuring that only active users can be removed.
.
The embeddings are generated locally in the DAG wit... | [
0.024570327252149582,
-0.005666167009621859,
-0.02140716463327408,
-0.03729425370693207,
0.06761057674884796,
-0.0414370559155941,
0.02552201598882675,
0.05578209087252617,
0.00817415863275528,
-0.0056661502458155155,
0.012319332920014858,
0.015086905099451542,
0.027492491528391838,
-0.052... |
b692999d-15d3-500e-b312-c64b6bbf0a35 | /course/beginners/module-5/analyst-queries/ | Analyst Queries | Going Further: Cross-Language Comparison | course | false | Analyst Queries > Going Further: Cross-Language Comparison
Supply chain news often appears in Japanese, Mandarin, Korean, or Vietnamese before it reaches an English wire, and `all-MiniLM-L6-v2` cannot read any of it. Reaching those sources is one substitution: point `DENSE_MODEL` at a multilingual model such as `intfl... | [
0.048971857875585556,
-0.02857392095029354,
-0.07303258031606674,
-0.01765100285410881,
0.07975658029317856,
-0.03482145816087723,
0.057817492634058,
0.03522058576345444,
0.06361444294452667,
-0.023810381069779396,
0.023040268570184708,
0.004419179167598486,
0.06184173747897148,
0.00614802... |
b697e875-2c43-5b9d-b424-35a964b7f5b6 | /articles/hyperbolic-embeddings-qdrant/ | Why Hyperbolic Embeddings Matter for Hierarchical Data | Testing Hyperbolic Embeddings | article | false | Why Hyperbolic Embeddings Matter for Hierarchical Data > Testing Hyperbolic Embeddings
So we would not treat this as a universal win. We would treat it as a strong reason to test hyperbolic embeddings when the data itself is hierarchical. | [
0.043356187641620636,
-0.006114459130913019,
-0.041125278919935226,
-0.07136008143424988,
0.03355563431978226,
-0.04040020704269409,
0.08152039349079132,
0.03803737834095955,
0.07144807279109955,
0.027349408715963364,
0.05059454217553139,
0.02840338833630085,
0.09944715350866318,
-0.031116... |
b6ce54be-5d84-5899-bd70-3ceb7be289c7 | /documentation/search/filtering/ | Filtering | Filtering Conditions > Range | documentation | true | Filtering > Filtering Conditions > Range
```python
models.FieldCondition(
key="price",
range=models.Range(
gt=None,
gte=100.0,
lt=None,
lte=450.0,
),
)
```
The `range` condition sets the range of possible values for stored payload values.
If several values are stored, at le... | [
-0.02359677106142044,
-0.04274807870388031,
-0.037459757179021835,
-0.02863307110965252,
0.05449913814663887,
-0.02678535506129265,
0.03076881915330887,
0.05728346109390259,
0.08101549744606018,
-0.03649093955755234,
0.03050951287150383,
-0.005967872682958841,
0.06752171367406845,
-0.02570... |
b6e2f823-37f1-53e7-8a2d-a56032a85df2 | /documentation/examples/graphrag-qdrant-neo4j/ | GraphRAG with Qdrant and Neo4j | Implementation > Prerequisites > Neo4j Setup | documentation | true | GraphRAG with Qdrant and Neo4j > Implementation > Prerequisites > Neo4j Setup
You can use [Neo4j Aura](https://neo4j.com/product/auradb/), Neo4j's managed service, or host Neo4j yourself. This tutorial uses Neo4j Aura:
- Go to the [Neo4j Aura console](https://console.neo4j.io/) and sign up or log in.
- After setting ... | [
0.019149087369441986,
-0.0678531676530838,
-0.023446720093488693,
-0.02294332906603813,
0.07087443023920059,
-0.05147450417280197,
0.036753058433532715,
0.06094158813357353,
0.011258958838880062,
-0.04068037122488022,
0.03584771603345871,
0.011509939096868038,
0.06937745213508606,
-0.05755... |
b6e3c876-e2f2-577e-8825-3c006a70ff62 | /documentation/migrate-to-qdrant/from-qdrant/ | From Qdrant | documentation | false | From Qdrant
Use the `qdrant` subcommand to copy a collection from one Qdrant instance to another or copy data between collections within the same instance.
If the target collection does not exist, the tool automatically creates it using the source collection's configuration (vector config, HNSW settings, quantization... | [
0.05922970548272133,
-0.06806104630231857,
-0.03385322913527489,
-0.035608433187007904,
0.0515524223446846,
-0.016126815229654312,
0.0912584662437439,
0.04262879118323326,
0.06049717962741852,
0.000054351210565073416,
0.01600753143429756,
0.006275518797338009,
0.07743679732084274,
-0.02219... | |
b6ee2411-b8c7-510a-b981-d22f57922600 | /documentation/frameworks/honeyhive/ | HoneyHive | Viewing Traces in HoneyHive | documentation | false | HoneyHive > Viewing Traces in HoneyHive
After running your RAG pipeline with Qdrant, you can view the traces in the HoneyHive UI:
1. Navigate to your project in the HoneyHive dashboard
2. Click on the "Traces" tab to see all the traces from your RAG pipeline
3. Click on a specific trace to see detailed information ab... | [
0.07063046097755432,
-0.05972330644726753,
-0.039062127470970154,
-0.06630951911211014,
0.03521260619163513,
-0.03657211735844612,
0.06136546656489372,
0.02827472612261772,
0.027765918523073196,
-0.05668927729129791,
-0.00029505707789212465,
0.029622334986925125,
0.08075795322656631,
-0.02... |
b6fc16d2-0a00-50a3-a60e-4544a291ceb4 | /documentation/ops-configuration/administration/ | Administration | Strict Mode > Maximum Number of Payload Index Count | documentation | true | Administration > Strict Mode > Maximum Number of Payload Index Count
Setting `max_payload_index_count` caps the maximum number of payload index that can exist on a collection.
```python
from qdrant_client import QdrantClient, models
client = QdrantClient(url="http://localhost:6333")
client.create_collection(
co... | [
-0.005470900796353817,
0.0038999775424599648,
-0.03138080984354019,
-0.019802656024694443,
0.0729852095246315,
-0.04382551461458206,
0.02437921054661274,
0.05234800651669502,
0.07971090823411942,
-0.07343113422393799,
0.016126243397593498,
0.016744866967201233,
0.03020942583680153,
-0.0684... |
b7205049-0b93-55c6-a37e-9219d8189f3a | /documentation/manage-data/payload/ | Payload | Update payload | documentation | true | Payload > Update payload
Updating payloads in Qdrant offers flexible methods to manage vector metadata. The **set payload** method updates specific fields while keeping others unchanged, while the **overwrite** method replaces the entire payload. Developers can also use **clear payload** to remove all metadata or dele... | [
0.051902808248996735,
-0.029720105230808258,
-0.03368741646409035,
-0.01509979460388422,
0.028603099286556244,
-0.020693207159638405,
0.03299962356686592,
0.026491403579711914,
0.06743886321783066,
-0.04192798584699631,
0.03237495943903923,
0.019639765843749046,
0.026069916784763336,
-0.01... |
b739f33d-c2ba-5f88-a682-ee6c8fe13518 | /documentation/ops-configuration/quotas/ | Resource Quotas | Monitoring Quotas | documentation | true | Resource Quotas > Monitoring Quotas
Each node reports whether it's currently at or over a limit through the `quota_exceeded` metric on [`/metrics`](/documentation/ops-monitoring/monitoring/#node-metrics-metrics). Memory and disk get their own series, because you free them with different actions:
```text
quota_exceede... | [
0.03276281803846359,
-0.03988661617040634,
-0.025172092020511627,
-0.03435174375772476,
0.03503425791859627,
-0.013618625700473785,
0.042048484086990356,
0.06117331609129906,
0.08752065151929855,
-0.061048638075590134,
0.029745563864707947,
-0.017171308398246765,
0.061680667102336884,
-0.0... |
b74b49cb-97c9-52fe-bc18-45c3b562ec9e | /documentation/faq/qdrant-fundamentals/ | Qdrant Fundamentals | Search > What are the default rescore values for different quantization methods? | documentation | false | Qdrant Fundamentals > Search > What are the default rescore values for different quantization methods?
By default, rescoring is only enabled for four quantization methods: binary quantization, TurboQuant 1 Bit, TurboQuant 1.5 Bit, and TurboQuant 2 Bit. Other quantization methods do not rescore by default. The default ... | [
0.05036841332912445,
-0.02319193258881569,
-0.013540772721171379,
-0.018988830968737602,
0.05062467232346535,
-0.03817814588546753,
0.035655826330184937,
0.0371738001704216,
0.09595934301614761,
0.012446019798517227,
0.0570809543132782,
0.019454892724752426,
0.02810007892549038,
-0.0070203... |
b750233b-5e72-5d1a-9f9c-0b074df2a7f2 | /articles/minicoil/ | miniCOIL: on the Road to Usable Sparse Neural Retrieval | The Good, the Bad and the Ugly > Sparse Neural Retrieval | article | false | miniCOIL: on the Road to Usable Sparse Neural Retrieval > The Good, the Bad and the Ugly > Sparse Neural Retrieval
So, on one side, we have weak control over matching, sometimes leading to too broad retrieval results, and on the other—lightweight, explainable and fast term-based retrievers like BM25, incapable of capt... | [
0.052007123827934265,
-0.03811638057231903,
-0.03719707950949669,
-0.04044314846396446,
0.05345718562602997,
-0.04658279940485954,
0.04796399176120758,
0.017819683998823166,
0.07112444192171097,
-0.019846880808472633,
0.024736851453781128,
0.021654104813933372,
0.08578398078680038,
-0.0105... |
b7544a06-b8a5-535d-b9ab-664c875e6d00 | /documentation/private-cloud/qdrant-cluster-management/ | Managing a Cluster | Authentication and Authorization > Configuring TLS for Database Access | documentation | true | Managing a Cluster > Authentication and Authorization > Configuring TLS for Database Access
If you want to configure TLS for accessing your Qdrant database, there are two options:
* You can offload TLS at the ingress or loadbalancer level.
* You can configure TLS directly in the Qdrant database.
If you want to confi... | [
0.036962058395147324,
-0.04019930586218834,
-0.03852196782827377,
-0.031034374609589577,
0.0379815511405468,
-0.05980939418077469,
0.0712672621011734,
0.03040061704814434,
0.04995308443903923,
-0.015374680981040001,
0.03911425173282623,
-0.008395316079258919,
0.08329565078020096,
-0.003122... |
b75ab755-dd43-5010-94c0-733a6b1eb1cb | /documentation/manage-data/indexing/ | Indexing | Full-Text Index > Tokenizers | documentation | false | Indexing > Full-Text Index > Tokenizers
Tokenizers are algorithms used to split text into smaller units called tokens, which are then indexed and searched in a full-text index.
In the context of Qdrant, tokenizers determine how string payloads are broken down for efficient filtering.
The choice of tokenizer affects ho... | [
0.026935428380966187,
-0.01236524898558855,
-0.04714220389723778,
-0.07588791847229004,
0.054456692188978195,
-0.01825663261115551,
0.04715722054243088,
0.03379921242594719,
0.05138438567519188,
-0.011861390434205532,
0.06815863400697708,
0.018060481175780296,
0.027979951351881027,
-0.0043... |
b78f4119-7708-5720-ad8a-e6e05e728451 | /articles/late-interaction-models/ | Late Interaction Retrieval with Dense Token Embeddings | Setting Up Qdrant for Late Interaction | article | true | Late Interaction Retrieval with Dense Token Embeddings > Setting Up Qdrant for Late Interaction
The new Query API in Qdrant 1.10 enables the construction of even more complex retrieval pipelines. We can use the single vector created after pooling for the initial retrieval step and then rerank the results using the out... | [
0.050982482731342316,
-0.04745236784219742,
-0.01616872288286686,
-0.04221692308783531,
0.06741128861904144,
-0.07265520095825195,
0.05579870566725731,
0.06451451778411865,
0.09573553502559662,
-0.038065649569034576,
0.02727496065199375,
0.029124677181243896,
0.05027727782726288,
-0.026496... |
b791b26d-1bd8-51ef-a3c1-d6977aa0d774 | /documentation/scaling/distributed_deployment/ | Distributed Deployment | Sharding > Shard Transfer Method | documentation | false | Distributed Deployment > Sharding > Shard Transfer Method
| Method: | Stream records | Snapshot | WAL delta |
|:---|:---|:---|:---|
| **Version** | v0.8.0+ | v1.7.0+ | v1.8.0+ |
| **Target** | New/existing shard | New/existing shard | Existing shard |
| **Connectivity** | Internal gRPC API (6335) | REST API (6333)Inte... | [
0.007743156515061855,
-0.02880408987402916,
-0.06811460852622986,
-0.008693072013556957,
0.06458356976509094,
-0.010962391272187233,
0.04102825000882149,
0.042675331234931946,
0.07000477612018585,
-0.024010485038161278,
0.028078118339180946,
0.044503334909677505,
0.0064954995177686214,
-0.... |
b7a4ef40-1db8-5eaa-88e8-7a7c6ebbc03c | /documentation/migration-guidance/pre-migration-baseline/ | Pre-Migration Baseline | Output | documentation | false | Pre-Migration Baseline > Output
After completing this step, you should have four artifacts:
1. **Collection inventory** (JSON or YAML): names, counts, dimensions, metrics, index params
2. **Metadata sample** (JSONL): representative records with all fields and types
3. **Baseline queries** (JSON): query vectors, filte... | [
0.04513059929013252,
-0.03145727142691612,
-0.009667209349572659,
-0.03623401001095772,
0.047351524233818054,
-0.030961666256189346,
0.04548923671245575,
0.01831248588860035,
0.054754190146923065,
-0.04729025438427925,
0.04349266737699509,
0.0052262600511312485,
0.06344648450613022,
-0.035... |
b7aae85b-053f-5ca5-881c-29eeebaedd73 | /documentation/migration-guidance/ | Migration Guidance | The Four Stages of Verification | documentation | false | Migration Guidance > The Four Stages of Verification
This guide proposes four distinct stages of verification, each building on the next to ensure a successful migration:
1. **[Pre-Migration Baseline](/documentation/migration-guidance/pre-migration-baseline/):** Before starting the migration, capture what "correct" l... | [
0.03221820667386055,
-0.03845857456326485,
-0.045874468982219696,
-0.038238275796175,
0.044399015605449677,
-0.016628090292215347,
0.03398972749710083,
0.03653138875961304,
0.09572075307369232,
-0.03420363366603851,
0.01118021085858345,
0.004611388314515352,
0.03754165768623352,
-0.0270485... |
b7aca223-4f8a-59b4-963c-eb453579b8a7 | /documentation/ops-monitoring/monitoring/ | Monitoring & Telemetry | Kubernetes Health Endpoints | documentation | false | Monitoring & Telemetry > Kubernetes Health Endpoints
*Available as of v1.5.0*
Qdrant exposes three endpoints, namely
[`/healthz`](http://localhost:6333/healthz),
[`/livez`](http://localhost:6333/livez) and
[`/readyz`](http://localhost:6333/readyz), to indicate the current status of the
Qdrant server.
These currently... | [
0.03050612285733223,
-0.021603407338261604,
-0.02320624329149723,
-0.06238625571131706,
0.044869232922792435,
-0.05352938920259476,
0.07559249550104141,
0.022427961230278015,
0.06434869021177292,
-0.02530542016029358,
0.024402566254138947,
0.0389062836766243,
0.03819369152188301,
-0.057997... |
b7bb91a4-7edf-5577-ac0b-59c6a42e4f92 | /course/beginners/module-5/knowledge-check/ | Knowledge Check | course | false | Knowledge Check
Because two of the sources are not new modalities. A transcript is text the moment it has been transcribed, and a video frame is an image the moment it has been sampled, so both reuse spaces that already exist. Adding a separate vector for transcripts would mean two named vectors holding the same 384-d... | [
0.03897522762417793,
-0.014090248383581638,
-0.08790462464094162,
-0.039128776639699936,
0.06428084522485733,
-0.021904518827795982,
0.06640877574682236,
0.03136321157217026,
0.051128268241882324,
-0.0291642714291811,
0.026241235435009003,
0.01306881383061409,
0.043363217264413834,
-0.0109... | |
b7e90ed6-91aa-528e-b63e-a9274d3d136f | /documentation/edge/edge-api/snapshots/ | Snapshots | snapshot_manifest | documentation | true | Snapshots > snapshot_manifest
Returns the shard's snapshot manifest, which describes its segments and their metadata. Pass it to a server when requesting a partial snapshot so the server sends only the segments that have changed.
```python
def snapshot_manifest(self) -> Any
```
**Returns** a JSON-like value in Pytho... | [
-0.005769685376435518,
-0.026474982500076294,
-0.018692074343562126,
-0.027903730049729347,
0.002067266032099724,
-0.014338160865008831,
0.0713317021727562,
0.030151275917887688,
0.05672382190823555,
-0.008862251415848732,
0.06407979875802994,
0.037397902458906174,
0.052488502115011215,
-0... |
b7e9e991-2731-5ddd-b7e8-51d9f687658b | /documentation/tutorials-operations/gpu-accelerated-hnsw-indexing/ | GPU-Accelerated HNSW Indexing | Analyzing the Results | documentation | true | GPU-Accelerated HNSW Indexing > Analyzing the Results
### HNSW Indexing Time
Parse `optimizations.jsonl` and retrieve the starting and end time for the optimizations, then compute the total time between them.
```python
def hnsw_indexing_time(optimizations_file: str) -> dict:
with open(optimizations_file) as f:
... | [
0.009127291850745678,
-0.042632751166820526,
-0.036734290421009064,
-0.03191620483994484,
0.06846990436315536,
-0.02386835217475891,
0.03581281006336212,
0.027806157246232033,
0.0804063007235527,
-0.012879958376288414,
0.06099161505699158,
0.018825797364115715,
0.06293953955173492,
-0.0308... |
b800cbdc-5551-5f97-a24d-5386bce4a9f1 | /course/essentials/day-0/pitstop-project/ | Project: Building Your First Vector Search System | Troubleshooting | course | false | Project: Building Your First Vector Search System > Troubleshooting
**No results from filtered search?** If no vectors satisfy the filter conditions, Qdrant returns an empty result set. Try adjusting your filter values or checking your payload data.
**Collection already exists?** Use `client.delete_collection(collect... | [
0.06591218709945679,
-0.04951101541519165,
-0.03307352960109711,
-0.008718490600585938,
0.07221973687410355,
-0.023692723363637924,
0.05398556962609291,
0.02301534079015255,
0.07480118423700333,
-0.055457502603530884,
0.01504471804946661,
0.03856454789638519,
0.06934922933578491,
-0.044623... |
b80b6140-d54f-54f5-8947-191585426e0c | /articles/what-is-vector-quantization/ | What is Vector Quantization? | Rescoring, Oversampling, and Reranking | article | false | What is Vector Quantization? > Rescoring, Oversampling, and Reranking
When we use quantization methods like Scalar, Binary, or Product Quantization, we're compressing our vectors to save memory and improve performance. However, this compression removes some detail from the original vectors.
This can slightly reduce t... | [
0.06278058886528015,
-0.05938548967242241,
-0.049493495374917984,
-0.05144490301609039,
0.07699646800756454,
0.00037165856338106096,
0.04991592466831207,
0.0047806925140321255,
0.10508853942155838,
-0.036210644990205765,
0.04252665489912033,
0.013442701660096645,
0.06273350864648819,
-0.01... |
b813f58c-eca3-5101-a8ed-f25e4960a100 | /documentation/production-checklist/ | Production Checklist | 3. Quantization | documentation | false | Production Checklist > 3. Quantization
Compress vectors to reduce memory footprint. [Quantization](/documentation/manage-data/quantization/) is one of the most impactful changes you can make before going to production.
- **Quantization** reduces the memory footprint of vectors, by compressing them to fewer bits. This... | [
0.03604603186249733,
-0.04650035873055458,
-0.043125201016664505,
-0.07125674188137054,
0.035812150686979294,
-0.0015744316624477506,
0.0645567998290062,
0.029158221557736397,
0.11816595494747162,
-0.04978834092617035,
0.019570063799619675,
-0.0006764768040739,
0.04151533544063568,
-0.0206... |
b82ade0a-6b22-541e-9cf6-0d6b15836971 | /course/essentials/day-7/haystack/ | Integrating with Haystack | Haystack Movie Recommendation Assistant > Real-World Applications | course | false | Integrating with Haystack > Haystack Movie Recommendation Assistant > Real-World Applications
This architecture extends beyond movie recommendations to various domains:
- **E-commerce**: Product recommendations with complex attribute filtering
- **Content Discovery**: Finding relevant articles, videos, or resources
-... | [
0.013553743250668049,
-0.05700141564011574,
-0.07612482458353043,
-0.00860822293907404,
0.06971559673547745,
-0.04094984382390976,
0.026422787457704544,
0.028496596962213516,
0.05583067238330841,
0.006174126174300909,
0.01319896150380373,
0.012639489024877548,
0.07596153765916824,
-0.02693... |
b87373ac-c737-55ab-8357-97ab7cdfed0b | /documentation/search/filtering/ | Filtering | Filtering Conditions > Exact Match Conditions > Match Except | documentation | true | Filtering > Filtering Conditions > Exact Match Conditions > Match Except
*Available as of v1.2.0*
In case you want to check if the stored value is not one of multiple values, you can use the Match Except condition.
Match Except works as a logical NOR for the given values.
It can also be described as a `NOT IN` operat... | [
0.01160842552781105,
-0.035877130925655365,
-0.0491027869284153,
-0.026279475539922714,
0.046887148171663284,
-0.005120361223816872,
0.024476101621985435,
0.01652100495994091,
0.05762103572487831,
-0.012352636083960533,
0.017446042969822884,
0.03616953641176224,
0.04578359052538872,
-0.030... |
b874536a-b4a5-5468-8836-14f2ac09a8c1 | /articles/late-interaction-models/ | Late Interaction Retrieval with Dense Token Embeddings | Practical Application: Enhancing Retrieval with Dense Models > Single Model Retrieval and Reranking Benchmarks | article | false | Late Interaction Retrieval with Dense Token Embeddings > Practical Application: Enhancing Retrieval with Dense Models > Single Model Retrieval and Reranking Benchmarks
Overall, adding a reranking step using the same model typically improves retrieval quality. However, the quality of various late interaction models is ... | [
0.04263335466384888,
-0.05325806513428688,
-0.030599970370531082,
-0.0631561428308487,
0.07192987948656082,
-0.07456362992525101,
0.0399276465177536,
0.06653384864330292,
0.08466725051403046,
-0.008264107629656792,
0.0538640096783638,
0.042860791087150574,
0.029104582965373993,
-0.03544428... |
b87bd7c7-de25-5a17-a6b0-c37d231203e2 | /course/essentials/day-3/sparse-vectors/ | Sparse Vectors and Inverted Indexes | Sparse Vectors in Qdrant > Create a Collection with Sparse Vectors | course | true | Sparse Vectors and Inverted Indexes > Sparse Vectors in Qdrant > Create a Collection with Sparse Vectors
```python
# Create collection with a named sparse vector
client.create_collection(
collection_name=<COLLECTION_NAME>,
sparse_vectors_config={
<SPARSE_VECTOR_NAME>: models.SparseVectorParams()
},... | [
0.05109401419758797,
-0.0602693185210228,
-0.016294226050376892,
-0.010036872699856758,
0.04647661745548248,
-0.014256768859922886,
0.02376376837491989,
0.0141370240598917,
0.09673383831977844,
-0.042871829122304916,
0.007205379195511341,
0.052810683846473694,
0.02983812801539898,
-0.02336... |
b87fd242-ede1-5cb9-842a-91f692538fa2 | /documentation/support/ | Support | Qdrant Cloud Support > Priority & SLA | documentation | false | Support > Qdrant Cloud Support > Priority & SLA
You will also be asked to select a **severity level**, which determines how your ticket is prioritized. The severity should reflect the impact on your system or customers.
- **Severity 1** – Critical impact: full service outage or data loss
- **Severity 2** – Major impa... | [
0.02939443662762642,
-0.010498491115868092,
-0.03165626525878906,
-0.02670644409954548,
0.053379856050014496,
-0.045674581080675125,
0.04417280852794647,
0.041362617164850235,
0.08414330333471298,
-0.036285944283008575,
0.005076538305729628,
-0.0005359759088605642,
0.03916445001959801,
-0.... |
b8805cbd-499b-552c-b122-2674899b0546 | /documentation/tutorials-build-essentials/data-ingestion-beginners/ | S3 Ingestion | Step 6: Verify the Ingestion | documentation | true | S3 Ingestion > Step 6: Verify the Ingestion
Check that every product arrived:
```python
print(client.count(COLLECTION, exact=True).count)
```
The output is `5` if you used the sample data.
You can also browse the collection in the [Qdrant Web UI](/documentation/web-ui/). For a Qdrant Cloud cluster, open your cluste... | [
0.06662757694721222,
-0.02576812170445919,
0.003858414478600025,
-0.04998955875635147,
0.07508328557014465,
-0.03313634172081947,
0.03914739564061165,
0.05819355696439743,
0.07486455887556076,
-0.03664277493953705,
0.012838923372328281,
0.04607372358441353,
0.039298463612794876,
-0.0344196... |
b888e4e5-219d-5e31-8884-0d9b83fcd9a3 | /documentation/ops-configuration/memory-tiers/ | Memory Tiers | Optimizing for Disk-Based Retrieval > Quantization | documentation | true | Memory Tiers > Optimizing for Disk-Based Retrieval > Quantization
[Quantization](/documentation/manage-data/quantization/) compresses vectors into a smaller representation. The quantized copy fits comfortably in RAM even when the original vectors are `cold`. This enables Qdrant to score most candidates against the qua... | [
0.05021481215953827,
-0.0701250359416008,
-0.026949886232614517,
-0.040026698261499405,
0.06895438581705093,
-0.01542259007692337,
0.051679857075214386,
0.007505223620682955,
0.10458900034427643,
-0.02490686997771263,
0.029965680092573166,
0.021718934178352356,
0.07103970646858215,
-0.0161... |
b8a6051a-7d87-5a01-8c32-e4925b0c3799 | /documentation/tutorials-search-engineering/using-relevance-feedback/ | Relevance Feedback Retrieval in Qdrant | Relevance Feedback | documentation | false | Relevance Feedback Retrieval in Qdrant > Relevance Feedback
> Relevance feedback distills signals about the relevance of current search results into the next retrieval iteration, surfacing better results over time.
The Relevance Feedback Query uses a small amount of model-generated feedback on the search results to g... | [
0.044507745653390884,
-0.04077756032347679,
-0.023537321016192436,
-0.04486069083213806,
0.054318878799676895,
-0.03590743616223335,
0.05063316971063614,
0.045775968581438065,
0.06020306795835495,
-0.010896719992160797,
0.017497379332780838,
0.03316042572259903,
0.09239494055509567,
-0.020... |
b8e16a7e-49aa-545f-8e73-c2b3992c8ec2 | /documentation/migration-guidance/search-quality/ | Search Quality | Tier 2: Statistical Sampling (Recommended) > Interpreting Tier 2 Results | documentation | true | Search Quality > Tier 2: Statistical Sampling (Recommended) > Interpreting Tier 2 Results
```py
def tier2_report(recalls, stats):
"""Print Tier 2 evaluation summary."""
print(f"Recall@{stats['k']} across {stats['num_queries']} queries:")
print(f" Mean: {stats['mean_recall']:.3f}")
print(f" Median: ... | [
0.01572074368596077,
-0.020465414971113205,
-0.04456639289855957,
-0.0274954903870821,
0.054018862545490265,
-0.004020964726805687,
0.03495291620492935,
0.03223998844623566,
0.08634353429079056,
-0.0037084969226270914,
0.039859380573034286,
0.032961342483758926,
0.06128833070397377,
-0.040... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.