From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Milvus 2.3.0 has 96 facts recorded in Dontopedia across 7 references, with 9 live disagreements.
Mostly:rdf:type(9), inverse of(5), has performance metric(5)
hasVersionNumberinverseOfhasPerformanceMetrichasHigherThroughputThanhasLowerStorageSizeThanrdfs:labelthroughputscalabilityOther subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. “X motherOf this subject” — and answer questions the forward facts can't. Grouped by predicate.
inverseOfInverse of(5)ex:concurrency-supportex:f1-scoreex:query-latencyex:scalabilityex:throughputhasLowerThroughputThanHas Lower Throughput Than(3)ex:Annoy-1.18.0ex:Faiss-1.7.3ex:Hnswlib-0.9.2measuredForMeasured for(2)ex:community_supportex:throughputusesUses(2)ex:search-functionex:User-Turn-2214containsLibraryContains Library(1)ex:matrix-data-structurecurrently-usesCurrently Uses(1)ex:User-Turn-2214hasHigherThroughputThanHas Higher Throughput Than(1)ex:Qdrant-0.8.1hasLowerEaseOfIntegrationThanHas Lower Ease of Integration Than(1)ex:Annoy-1.18.0hasLowerPrecisionRateThanHas Lower Precision Rate Than(1)ex:Annoy-1.18.0hasLowerRecallRateThanHas Lower Recall Rate Than(1)ex:Annoy-1.18.0hasLowerUptimeThanHas Lower Uptime Than(1)ex:Annoy-1.18.0hasMemberHas Member(1)ex:vector-search-librarieshasRowHas Row(1)ex:performance-matrixisPrecisionRateOfIs Precision Rate of(1)ex:0.96isRecallRateOfIs Recall Rate of(1)ex:0.95isScalabilityOfIs Scalability of(1)ex:0.9isStorageSizeOfIs Storage Size of(1)ex:1000notMeasuredForNot Measured for(1)ex:memory_usageonlyMeasuredForOnly Measured for(1)ex:scalabilitystorageSizeExceedsStorage Size Exceeds(1)ex:Annoy-1.18.0targetsVersionTargets Version(1)ex:Milvus-optimization-guideuses-currentlyUses Currently(1)ex:User-Turn-2214The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.
| Predicate | Value | Ref |
|---|---|---|
| Row Order | 1 | [2] |
| Has Unique Throughput Value | 1000 | [2] |
| Shares Identical Metrics With | Qdrant 0.8.1 | [2] |
| Is Only System With Throughput Data | true | [2] |
| Shares Scalability With | Qdrant 0.8.1 | [2] |
| Has Lowest Query Latency | 200 | [2] |
| Has Highest F1 Score | Qdrant 0.8.1 | [2] |
| Concurrency Support | 0.9 | [2] |
| Query Latency | 200 | [2] |
| F1 Score | 0.935 | [2] |
| Is Used by | User Turn 2214 | [3] |
| Has Throughput | High Throughput | [3] |
| Has Search Time | Low Search Time | [3] |
| Has F1 Score | F1 Scores | [3] |
| Has Precision | Superior Precision | [3] |
| Has Recall | Superior Recall | [3] |
| Has Higher Community Support Than | Faiss 1.7.3 | [1] |
| Has Community Support | 0.9 | [1] |
| Has Lower Throughput Than | Qdrant 0.8.1 | [1] |
| Has Highest Community Support | true | [1] |
| Has Highest Ease of Integration | true | [1] |
| Has Highest Uptime | true | [1] |
| Community Support | 0.9 | [1] |
| Ease of Integration | 0.9 | [1] |
| Uptime | 0.999 | [1] |
| Storage Size Is Smallest | true | [4] |
| Scalability Is Higher | true | [4] |
| Scalability Value | 0.9 | [4] |
| Tied for Highest Precision Rate | true | [4] |
| Tied for Highest Recall Rate | true | [4] |
| Has Lowest Storage Size Value | 1000 | [4] |
| Has Smallest Storage Size Difference | true | [4] |
| Has Scalability Metric | true | [4] |
| Has Higher Scalability Than | Faiss 1.7.3 | [4] |
| Has Highest Precision Rate | true | [4] |
| Has Highest Recall Rate | true | [4] |
| Has Smallest Storage Size Value | 1000 | [4] |
| Has Smallest Storage Size | true | [4] |
| Precision Rate | 0.96 | [4] |
| Recall Rate | 0.95 | [4] |
| Storage Size | 1000 | [4] |
| Has Memory Usage | 500 | [5] |
| Has Indexing Time | 300 | [5] |
| Has Query Latency | 200 | [5] |
| Has Index Size | 1000 | [5] |
| Is Only Library With Scalability Metric | true | [6] |
| Has Lowest Storage Size | true | [6] |
| Is Member of | Vector Search Libraries | [6] |
| Has Scalability | 0.9 | [6] |
| Has Precision Rate | 0.96 | [6] |
| Has Recall Rate | 0.95 | [6] |
| Has Storage Size | 1000 | [6] |
Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.
doc:beam/144b6238-dbb6-458e-99d6-f284a5160b1fmatrix.loc['Hnswlib 0.9.2', 'concurrency_support'] = 0.85 matrix.loc['Qdrant 0.8.1', 'concurrency_support'] = 0.9 matrix.loc['Weaviate 1.14.0', 'concurrency_support'] = 0.85 matrix.loc['Milvus 2.3.0', 'throughput'] = 1000 matrix.loc['Faiss…
doc:beam/98bc9425-2e1a-436c-9385-948ebc2769f1matrix.loc['Annoy 1.18.0', 'f1_score'] = 0.935 matrix.loc['Hnswlib 0.9.2', 'f1_score'] = 0.945 matrix.loc['Qdrant 0.8.1', 'f1_score'] = 0.955 matrix.loc['Weaviate 1.19.0', 'f1_score'] = 0.945 matrix.loc['Milvus 2.3.0', 'query_latency'] = 2…
doc:beam/854895db-e17a-401e-917b-ddd3a3b97e12Based on the current data, Milvus 2.3.0 and Qdrant 0.8.1 appear to be the best choices due to their superior recall, precision, and F1 scores, along with low search time and high throughput. Further evaluation of other metrics such as scala…
doc:beam/af22a9fe-7b7b-4f1b-a277-492b716acdedmatrix.loc['Hnswlib 0.9.2', 'search_time'] = 220 matrix.loc['Qdrant 0.8.1', 'search_time'] = 190 matrix.loc['Weaviate 1.14.0', 'search_time'] = 210 matrix.loc['Milvus 2.3.0', 'index_size'] = 1000 matrix.loc['Faiss 1.7.3', 'index_size'] = 1…
doc:beam/7962136c-c338-4cc2-87ff-eaf945be2841matrix.loc['Annoy 1.18.0', 'memory_usage'] = 600 matrix.loc['Hnswlib 0.9.2', 'memory_usage'] = 580 matrix.loc['Qdrant 0.8.1', 'memory_usage'] = 520 matrix.loc['Weaviate 1.14.0', 'memory_usage'] = 560 matrix.loc['Milvus 2.3.0', 'storage_siz…
doc:beam/dc4e867f-2dc3-4866-a506-665fdbdd3a9e'metric_type': 'L2' } client.create_index(collection_name, field_name='vector', index_params=index_params) # Insert some vectors vectors = [[1.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 8.0, 9.0]] ids = [1, 2, 3] client.insert(collection_nam…
Dontopedia is in a read-only public launch. Follow the references and disputed branches now; contributions will open after durable identity and moderation are in place.