Every Claude Code session starts from zero, no memory of previous sessions. This server uses mem0ai as a library and exposes 11 MCP tools for storing, searching, and managing memories. Qdrant handles vector storage, Ollama runs embeddings locally (bge-m3), and Neo4j optionally builds a knowledge graph.
Some engineering details HN might find interesting:
- Zero-config auth: auto-reads Claude Code's OAT token from ~/.claude/.credentials.json, detects token type (OAT vs API key), and configures the SDK accordingly. No separate API key needed.\n- Graph LLM ops (3 calls per add_memory) can be routed to Ollama (free/local), Gemini 2.5 Flash Lite (near-free), or a split-model where Gemini handles entity extraction (85.4% accuracy) and Claude handles contradiction detection (100% accuracy).
Python, MIT licensed, one-command install via uvx.
https://github.com/elvismdev/mem0-mcp-selfhosted"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Persistent memory for Claude Code with self-hosted Qdrant and Ollama"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/elvismdev/mem0-mcp-selfhosted"}},"_tags":["story","author_elvismdev","story_47053534","show_hn"],"author":"elvismdev","created_at":"2026-02-17T21:21:25Z","created_at_i":1771363285,"num_comments":0,"objectID":"47053534","points":8,"story_id":47053534,"story_text":"I built an MCP server that gives Claude Code long-term memory across sessions, backed by infrastructure you control.
Every Claude Code session starts from zero, no memory of previous sessions. This server uses mem0ai as a library and exposes 11 MCP tools for storing, searching, and managing memories. Qdrant handles vector storage, Ollama runs embeddings locally (bge-m3), and Neo4j optionally builds a knowledge graph.
Some engineering details HN might find interesting:
- Zero-config auth: auto-reads Claude Code's OAT token from ~/.claude/.credentials.json, detects token type (OAT vs API key), and configures the SDK accordingly. No separate API key needed.\n- Graph LLM ops (3 calls per add_memory) can be routed to Ollama (free/local), Gemini 2.5 Flash Lite (near-free), or a split-model where Gemini handles entity extraction (85.4% accuracy) and Claude handles contradiction detection (100% accuracy).
Python, MIT licensed, one-command install via uvx.
https://github.com/elvismdev/mem0-mcp-selfhosted","title":"Show HN: Persistent memory for Claude Code with self-hosted Qdrant and Ollama","updated_at":"2026-04-06T02:35:19Z","url":"https://github.com/elvismdev/mem0-mcp-selfhosted"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"harporoeder"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Is ClickHouse Ready to Replace Your Specialized Vector Database (Qdrant)?"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://blog.arguflow.ai/posts/clickhouse-vs-vector-database-qdrant/"}},"_tags":["story","author_harporoeder","story_37340844"],"author":"harporoeder","created_at":"2023-08-31T17:35:57Z","created_at_i":1693503357,"num_comments":0,"objectID":"37340844","points":8,"story_id":37340844,"title":"Is ClickHouse Ready to Replace Your Specialized Vector Database (Qdrant)?","updated_at":"2024-09-20T15:00:46Z","url":"https://blog.arguflow.ai/posts/clickhouse-vs-vector-database-qdrant/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"softwaredoug"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"The Qdrant Output Connector"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://opencrawling.org/blog/introducing-qdrant-output-connector.html"}},"_tags":["story","author_softwaredoug","story_49301104"],"author":"softwaredoug","created_at":"2026-08-14T16:35:18Z","created_at_i":1786725318,"num_comments":0,"objectID":"49301104","points":7,"story_id":49301104,"title":"The Qdrant Output Connector","updated_at":"2026-08-14T18:56:08Z","url":"https://opencrawling.org/blog/introducing-qdrant-output-connector.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"thoughtfullyso"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant \u2013 The vector database has a hybrid cloud now"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://qdrant.tech/blog/hybrid-cloud/"}},"_tags":["story","author_thoughtfullyso","story_40051016"],"author":"thoughtfullyso","created_at":"2024-04-16T12:14:48Z","created_at_i":1713269688,"num_comments":0,"objectID":"40051016","points":7,"story_id":40051016,"title":"Qdrant \u2013 The vector database has a hybrid cloud now","updated_at":"2024-09-20T16:49:40Z","url":"https://qdrant.tech/blog/hybrid-cloud/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"yagizdegirmenci"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant: Vector similarity search engine with extended filtering support"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/qdrant/qdrant"}},"_tags":["story","author_yagizdegirmenci","story_27119545"],"author":"yagizdegirmenci","children":[27119911],"created_at":"2021-05-11T15:46:46Z","created_at_i":1620748006,"num_comments":2,"objectID":"27119545","points":6,"story_id":27119545,"title":"Qdrant: Vector similarity search engine with extended filtering support","updated_at":"2024-09-20T08:31:10Z","url":"https://github.com/qdrant/qdrant"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"sabrinaaquino"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"XAI released Grok-1 and forked Qdrant"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/xai-org"}},"_tags":["story","author_sabrinaaquino","story_39740478"],"author":"sabrinaaquino","children":[39740886],"created_at":"2024-03-18T04:16:21Z","created_at_i":1710735381,"num_comments":1,"objectID":"39740478","points":6,"story_id":39740478,"title":"XAI released Grok-1 and forked Qdrant","updated_at":"2024-09-20T16:37:26Z","url":"https://github.com/xai-org"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"xojoc"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant \u2013 Vector Search Engine"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://qdrant.tech/"}},"_tags":["story","author_xojoc","story_29941275"],"author":"xojoc","children":[29941412],"created_at":"2022-01-14T22:18:17Z","created_at_i":1642198697,"num_comments":1,"objectID":"29941275","points":6,"story_id":29941275,"title":"Qdrant \u2013 Vector Search Engine","updated_at":"2024-09-20T10:14:58Z","url":"https://qdrant.tech/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"migrx"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant Vault Secrets Engine Plugin"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/migrx-io/vault-plugin-secrets-qdrant"}},"_tags":["story","author_migrx","story_41224606"],"author":"migrx","children":[41224607],"created_at":"2024-08-12T14:14:23Z","created_at_i":1723472063,"num_comments":1,"objectID":"41224606","points":5,"story_id":41224606,"title":"Qdrant Vault Secrets Engine Plugin","updated_at":"2024-09-20T17:35:48Z","url":"https://github.com/migrx-io/vault-plugin-secrets-qdrant"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"TalktoCrystal"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Open source db platform for vector db engines, qdrant, milvus, weaviate all in 1"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/apecloud/kubeblocks/blob/main/docs/user_docs/kubeblocks-for-vector-database/manage-vector-databases.md"}},"_tags":["story","author_TalktoCrystal","story_38840179"],"author":"TalktoCrystal","children":[38840328],"created_at":"2024-01-02T10:31:20Z","created_at_i":1704191480,"num_comments":1,"objectID":"38840179","points":5,"story_id":38840179,"title":"Open source db platform for vector db engines, qdrant, milvus, weaviate all in 1","updated_at":"2024-09-20T16:02:11Z","url":"https://github.com/apecloud/kubeblocks/blob/main/docs/user_docs/kubeblocks-for-vector-database/manage-vector-databases.md"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"talboren"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"The disk that never woke up: The Elasticsearch and Qdrant saga"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://www.elastic.co/search-labs/blog/vector-search-benchmark-elasticsearch-qdrant"}},"_tags":["story","author_talboren","story_48904782"],"author":"talboren","created_at":"2026-07-14T10:50:39Z","created_at_i":1784026239,"num_comments":0,"objectID":"48904782","points":5,"story_id":48904782,"title":"The disk that never woke up: The Elasticsearch and Qdrant saga","updated_at":"2026-07-14T15:12:32Z","url":"https://www.elastic.co/search-labs/blog/vector-search-benchmark-elasticsearch-qdrant"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"anonentity_lc"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"For the past couple of months, I\u2019ve been building a tool that enables natural language search over large codebases using Tree-Sitter for syntax parsing and Qdrant for vector-based retrieval.\nhttps://app.repogram.com
### How It Works
- Tree-Sitter is used to parse syntax trees and extract high-quality vector embeddings of code. - These embeddings are stored in Qdrant, enabling fast similarity search across your entire repo. - A combination of re-ranking processes refine search results, producing highly relevant answers to code-related questions.
The results have been incredibly accurate, thanks to the quality and structured nature of the embeddings. I've been working through the TypeORM issue backlog with some excellent results.
Repogram doesn\u2019t just help you search code, it enables a shared, evolving knowledge base:
- Questions are converted into public, protected, or private threads that remain searchable by repo members. - Over time, this builds up tribal knowledge that acts as an interactive alternative to traditional documentation. - Threads can be optionally summarized into searchable pages, making it easier to navigate discussions.
### Getting Started
- Enter a GitHub repo URL or name. - Sync the repo to generate embeddings. - Query your codebase using natural language.
### Currently supported languages:
- TypeScript - Javascript - Rust - Python - Go - Java - PHP - Swift - Ruby
...with additional support for Markdown. I\u2019ll be adding support for more languages over the coming weeks.
To get started, log in with GitHub to start exploring. I\u2019ve already synced a few large open-source libraries like TypeORM, Fastify, and BullMQ, so feel free to experiment\u2014or sync your own projects with a fine-grained PAT via the UI.
It's still very much a work in progress but the UI is fully functional. If you hit any bugs, let me know and I\u2019ll patch them as quickly as possible. Would love to hear your thoughts! Feedback, ideas, and feature suggestions are all welcome."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Advanced Code Search with Tree-sitter AST and Qdrant Vector DB"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://app.repogram.com/"}},"_tags":["story","author_anonentity_lc","story_43331090","show_hn"],"author":"anonentity_lc","created_at":"2025-03-11T10:37:55Z","created_at_i":1741689475,"num_comments":0,"objectID":"43331090","points":5,"story_id":43331090,"story_text":"For the past couple of months, I\u2019ve been building a tool that enables natural language search over large codebases using Tree-Sitter for syntax parsing and Qdrant for vector-based retrieval.\nhttps://app.repogram.com
### How It Works
- Tree-Sitter is used to parse syntax trees and extract high-quality vector embeddings of code. - These embeddings are stored in Qdrant, enabling fast similarity search across your entire repo. - A combination of re-ranking processes refine search results, producing highly relevant answers to code-related questions.
The results have been incredibly accurate, thanks to the quality and structured nature of the embeddings. I've been working through the TypeORM issue backlog with some excellent results.
Repogram doesn\u2019t just help you search code, it enables a shared, evolving knowledge base:
- Questions are converted into public, protected, or private threads that remain searchable by repo members. - Over time, this builds up tribal knowledge that acts as an interactive alternative to traditional documentation. - Threads can be optionally summarized into searchable pages, making it easier to navigate discussions.
### Getting Started
- Enter a GitHub repo URL or name. - Sync the repo to generate embeddings. - Query your codebase using natural language.
### Currently supported languages:
- TypeScript - Javascript - Rust - Python - Go - Java - PHP - Swift - Ruby
...with additional support for Markdown. I\u2019ll be adding support for more languages over the coming weeks.
To get started, log in with GitHub to start exploring. I\u2019ve already synced a few large open-source libraries like TypeORM, Fastify, and BullMQ, so feel free to experiment\u2014or sync your own projects with a fine-grained PAT via the UI.
It's still very much a work in progress but the UI is fully functional. If you hit any bugs, let me know and I\u2019ll patch them as quickly as possible. Would love to hear your thoughts! Feedback, ideas, and feature suggestions are all welcome.","title":"Show HN: Advanced Code Search with Tree-sitter AST and Qdrant Vector DB","updated_at":"2025-03-11T15:40:10Z","url":"https://app.repogram.com/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"chelbi"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Hi HN!
I built a tool called Qdrant Vector Aggregator to solve a common problem in vector-DB workflows:\nYou store chunk embeddings in Qdrant, but later you want document-level embeddings without losing the full document text."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Qdrant Vector Aggregator"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/vinerya/qdrant_vector_aggregator"}},"_tags":["story","author_chelbi","story_46031237","show_hn"],"author":"chelbi","children":[46031583,46033232],"created_at":"2025-11-24T07:17:35Z","created_at_i":1763968655,"num_comments":4,"objectID":"46031237","points":4,"story_id":46031237,"story_text":"Hi HN!
I built a tool called Qdrant Vector Aggregator to solve a common problem in vector-DB workflows:\nYou store chunk embeddings in Qdrant, but later you want document-level embeddings without losing the full document text.","title":"Show HN: Qdrant Vector Aggregator","updated_at":"2026-04-02T19:23:34Z","url":"https://github.com/vinerya/qdrant_vector_aggregator"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"sabrinaaquino"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Chroma vs. Qdrant vs. Weaviate Benchmark"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.adesso.de/en/news/blog/procedure-for-the-creation-of-vector-databases-benchmark-tests-and-their-results.jsp"}},"_tags":["story","author_sabrinaaquino","story_39708365"],"author":"sabrinaaquino","children":[39709380,39709612],"created_at":"2024-03-14T20:05:24Z","created_at_i":1710446724,"num_comments":1,"objectID":"39708365","points":4,"story_id":39708365,"title":"Chroma vs. Qdrant vs. Weaviate Benchmark","updated_at":"2024-09-20T16:33:57Z","url":"https://www.adesso.de/en/news/blog/procedure-for-the-creation-of-vector-databases-benchmark-tests-and-their-results.jsp"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Alyka"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant v0.11: fully scalable vector search engine"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/qdrant/qdrant/releases/tag/v0.11.0"}},"_tags":["story","author_Alyka","story_33345341"],"author":"Alyka","children":[33345342],"created_at":"2022-10-26T15:37:12Z","created_at_i":1666798632,"num_comments":1,"objectID":"33345341","points":4,"story_id":33345341,"title":"Qdrant v0.11: fully scalable vector search engine","updated_at":"2024-09-20T12:24:01Z","url":"https://github.com/qdrant/qdrant/releases/tag/v0.11.0"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ahsekka"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Hi HN \u2014 sharing ragctl, an open-source CLI for the most failure-prone part of RAG pipelines: document ingestion, OCR, parsing/cleaning, and chunking.
Vector DB setup is fairly standardized now, but getting high-quality, consistent text + metadata into it still takes a lot of brittle glue code. ragctl aims to make that \u201cpre-vector\u201d step repeatable: turn messy documents into retrieval-ready chunks in a few commands.
Features\n \u2022 Multi-format input: PDF, DOCX, HTML, images\n \u2022 OCR for scanned/image-based docs\n \u2022 Semantic chunking (LangChain)\n \u2022 Batch runs with retries + error handling\n \u2022 Output: direct ingestion into Qdrant (for now)
Looking for feedback\n \u2022 DX: is the CLI intuitive?\n \u2022 Performance / edge cases: weird PDFs, mixed layouts, tables\n \u2022 Roadmap: which connectors (S3, Slack, Notion) or vector stores should be next?
Repo: https://github.com/datallmhub/ragstudio\nHappy to answer questions about the architecture and chunking approach."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Ragctl \u2013 document ingestion CLI for RAG (OCR, chunking, Qdrant)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/datallmhub/ragstudio"}},"_tags":["story","author_ahsekka","story_46371520","show_hn"],"author":"ahsekka","created_at":"2025-12-24T01:35:55Z","created_at_i":1766540155,"num_comments":0,"objectID":"46371520","points":4,"story_id":46371520,"story_text":"Hi HN \u2014 sharing ragctl, an open-source CLI for the most failure-prone part of RAG pipelines: document ingestion, OCR, parsing/cleaning, and chunking.
Vector DB setup is fairly standardized now, but getting high-quality, consistent text + metadata into it still takes a lot of brittle glue code. ragctl aims to make that \u201cpre-vector\u201d step repeatable: turn messy documents into retrieval-ready chunks in a few commands.
Features\n \u2022 Multi-format input: PDF, DOCX, HTML, images\n \u2022 OCR for scanned/image-based docs\n \u2022 Semantic chunking (LangChain)\n \u2022 Batch runs with retries + error handling\n \u2022 Output: direct ingestion into Qdrant (for now)
Looking for feedback\n \u2022 DX: is the CLI intuitive?\n \u2022 Performance / edge cases: weird PDFs, mixed layouts, tables\n \u2022 Roadmap: which connectors (S3, Slack, Notion) or vector stores should be next?
Repo: https://github.com/datallmhub/ragstudio\nHappy to answer questions about the architecture and chunking approach.","title":"Show HN: Ragctl \u2013 document ingestion CLI for RAG (OCR, chunking, Qdrant)","updated_at":"2026-03-05T23:17:28Z","url":"https://github.com/datallmhub/ragstudio"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tsenturk"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Implementing RAG from Scratch with Python, Qdrant, and Docling"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://techlife.blog/posts/implementing-rag-from-scratch-qdrant/"}},"_tags":["story","author_tsenturk","story_46088088"],"author":"tsenturk","created_at":"2025-11-29T15:07:08Z","created_at_i":1764428828,"num_comments":0,"objectID":"46088088","points":4,"story_id":46088088,"title":"Implementing RAG from Scratch with Python, Qdrant, and Docling","updated_at":"2026-03-05T23:10:13Z","url":"https://techlife.blog/posts/implementing-rag-from-scratch-qdrant/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"nix_95"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Deploying an AI agent often means dealing with backend setup, managing vector databases, and other tedious tasks. To simplify the process, we took a low-code approach using tools like Flowise and Qdrant.
Flowise made it easy to build the logic without heavy coding, while Qdrant handled vector storage seamlessly. Pre-configured environments and automation helped from Qubinets cut down on redundant steps. Here's a breakdown of how we streamlined the entire deployment process and got our AI agent running quickly."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Quick AI Agent Deployment with Flowise and Qdrant"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://qubinets.com/how-to-build-an-ai-agent-with-qubinets/"}},"_tags":["story","author_nix_95","story_41766152","show_hn"],"author":"nix_95","created_at":"2024-10-07T14:05:55Z","created_at_i":1728309955,"num_comments":0,"objectID":"41766152","points":4,"story_id":41766152,"story_text":"Deploying an AI agent often means dealing with backend setup, managing vector databases, and other tedious tasks. To simplify the process, we took a low-code approach using tools like Flowise and Qdrant.
Flowise made it easy to build the logic without heavy coding, while Qdrant handled vector storage seamlessly. Pre-configured environments and automation helped from Qubinets cut down on redundant steps. Here's a breakdown of how we streamlined the entire deployment process and got our AI agent running quickly.","title":"Show HN: Quick AI Agent Deployment with Flowise and Qdrant","updated_at":"2024-10-08T20:57:54Z","url":"https://qubinets.com/how-to-build-an-ai-agent-with-qubinets/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mtrofficus"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Playing Mario Kart 64 using (just) Qdrant"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://twitter.com/MTrofficus/status/1794507641156194370"}},"_tags":["story","author_mtrofficus","story_40492766"],"author":"mtrofficus","created_at":"2024-05-27T17:22:29Z","created_at_i":1716830549,"num_comments":0,"objectID":"40492766","points":4,"story_id":40492766,"title":"Playing Mario Kart 64 using (just) Qdrant","updated_at":"2024-09-20T17:09:12Z","url":"https://twitter.com/MTrofficus/status/1794507641156194370"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"gaocegege"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Pgvector vs. Qdrant"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://nirantk.com/writing/pgvector-vs-qdrant/"}},"_tags":["story","author_gaocegege","story_37305291"],"author":"gaocegege","created_at":"2023-08-29T09:39:36Z","created_at_i":1693301976,"num_comments":0,"objectID":"37305291","points":4,"story_id":37305291,"title":"Pgvector vs. Qdrant","updated_at":"2024-09-20T14:56:44Z","url":"https://nirantk.com/writing/pgvector-vs-qdrant/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"randomsd"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant Open-Source Vector Similarity Search"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://read.theneedle.ai/p/qdrant-7-5m-open-source-vector-similarity-search"}},"_tags":["story","author_randomsd","story_35633681"],"author":"randomsd","created_at":"2023-04-19T21:43:37Z","created_at_i":1681940617,"num_comments":0,"objectID":"35633681","points":4,"story_id":35633681,"title":"Qdrant Open-Source Vector Similarity Search","updated_at":"2024-09-20T13:48:49Z","url":"https://read.theneedle.ai/p/qdrant-7-5m-open-source-vector-similarity-search"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"do-me"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"With just two main components you can create a fully working semantic search app. The client performs the heavy work of embedding calculation and simply sends the vector to Qdrant (vector database with built-in API) performing the actual search. Note that the model I'm using in the frontend is quantized (weighs only 30mb) and hence outputs slightly different results in comparison to the original model."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Semantic Search with Qdrant and Transformers.js"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://geo.rocks/post/qdrant-transformers-js-semantic-search/"}},"_tags":["story","author_do-me","story_35456569","show_hn"],"author":"do-me","created_at":"2023-04-05T16:18:41Z","created_at_i":1680711521,"num_comments":0,"objectID":"35456569","points":4,"story_id":35456569,"story_text":"With just two main components you can create a fully working semantic search app. The client performs the heavy work of embedding calculation and simply sends the vector to Qdrant (vector database with built-in API) performing the actual search. Note that the model I'm using in the frontend is quantized (weighs only 30mb) and hence outputs slightly different results in comparison to the original model.","title":"Show HN: Semantic Search with Qdrant and Transformers.js","updated_at":"2024-09-20T13:49:18Z","url":"https://geo.rocks/post/qdrant-transformers-js-semantic-search/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"danielsgriffin"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Quora Engineering on Qdrant: Building Embedding Search at Quora"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://quoraengineering.quora.com/Building-Embedding-Search-at-Quora"}},"_tags":["story","author_danielsgriffin","story_41630650"],"author":"danielsgriffin","children":[41630651,41640726],"created_at":"2024-09-23T21:16:29Z","created_at_i":1727126189,"num_comments":1,"objectID":"41630650","points":3,"story_id":41630650,"title":"Quora Engineering on Qdrant: Building Embedding Search at Quora","updated_at":"2024-09-24T20:45:31Z","url":"https://quoraengineering.quora.com/Building-Embedding-Search-at-Quora"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"timvisee"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant 1.8.0"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://qdrant.tech/articles/qdrant-1.8.x/"}},"_tags":["story","author_timvisee","story_39715336"],"author":"timvisee","children":[39716183],"created_at":"2024-03-15T13:38:29Z","created_at_i":1710509909,"num_comments":1,"objectID":"39715336","points":3,"story_id":39715336,"title":"Qdrant 1.8.0","updated_at":"2024-09-20T16:34:48Z","url":"https://qdrant.tech/articles/qdrant-1.8.x/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"andre-z"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant vector db v1.4 released with data visualization UI"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/qdrant/qdrant/releases/tag/v1.4.0"}},"_tags":["story","author_andre-z","story_36985348"],"author":"andre-z","children":[36985349],"created_at":"2023-08-03T13:05:05Z","created_at_i":1691067905,"num_comments":1,"objectID":"36985348","points":3,"story_id":36985348,"title":"Qdrant vector db v1.4 released with data visualization UI","updated_at":"2024-09-20T14:51:32Z","url":"https://github.com/qdrant/qdrant/releases/tag/v1.4.0"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tobacconcoffee"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Building Semantic Search in Rust with Qdrant and Shuttle (Live)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.youtube.com/watch?v=YLWSeiDh2o0"}},"_tags":["story","author_tobacconcoffee","story_36327906"],"author":"tobacconcoffee","children":[36328030],"created_at":"2023-06-14T15:59:48Z","created_at_i":1686758388,"num_comments":1,"objectID":"36327906","points":3,"story_id":36327906,"title":"Building Semantic Search in Rust with Qdrant and Shuttle (Live)","updated_at":"2024-09-20T14:21:54Z","url":"https://www.youtube.com/watch?v=YLWSeiDh2o0"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"avthar"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Postgres vs. Qdrant Vector Database Comparison"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://www.timescale.com/blog/pgvector-vs-qdrant"}},"_tags":["story","author_avthar","story_43832617"],"author":"avthar","created_at":"2025-04-29T13:51:21Z","created_at_i":1745934681,"num_comments":0,"objectID":"43832617","points":3,"story_id":43832617,"title":"Postgres vs. Qdrant Vector Database Comparison","updated_at":"2025-04-29T17:04:27Z","url":"https://www.timescale.com/blog/pgvector-vs-qdrant"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"nix_95"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"I\u2019ve been experimenting with Flowise and Qdrant to build AI agents, and I wanted to share a setup that eliminates the usual manual infrastructure deployment. Instead of manually doing cloud configuration and database setup, I used Qubinets to spin up everything in less than 5 minutes.
How I did it?
- I created a project in Qubinets (No cloud account needed for quick testing).\n- Selected Flowise AI (for building AI agents) and Qdrant (for vector storage).\n- Clicked Instantiate\u2014Qubinets handled the full deployment.\n- Waited a few minutes, then logged into Flowise and started building.
Why This is Useful
- No manual setup\u2014everything is pre-configured.\n- Great for prototyping\u2014test AI agents fast without worrying about infrastructure.\n- Scalable\u2014switch to a persistent cloud setup when moving to production.
Here's the full breakdown -> https://youtu.be/yEJyyjbHsPc?si=dTVF5jGVVWq5Wna5"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: A faster way to set up Flowise and Qdrant for AI agent development"}},"_tags":["story","author_nix_95","story_43159822","show_hn"],"author":"nix_95","created_at":"2025-02-24T14:20:20Z","created_at_i":1740406820,"num_comments":0,"objectID":"43159822","points":3,"story_id":43159822,"story_text":"I\u2019ve been experimenting with Flowise and Qdrant to build AI agents, and I wanted to share a setup that eliminates the usual manual infrastructure deployment. Instead of manually doing cloud configuration and database setup, I used Qubinets to spin up everything in less than 5 minutes.
How I did it?
- I created a project in Qubinets (No cloud account needed for quick testing).\n- Selected Flowise AI (for building AI agents) and Qdrant (for vector storage).\n- Clicked Instantiate\u2014Qubinets handled the full deployment.\n- Waited a few minutes, then logged into Flowise and started building.
Why This is Useful
- No manual setup\u2014everything is pre-configured.\n- Great for prototyping\u2014test AI agents fast without worrying about infrastructure.\n- Scalable\u2014switch to a persistent cloud setup when moving to production.
Here's the full breakdown -> https://youtu.be/yEJyyjbHsPc?si=dTVF5jGVVWq5Wna5","title":"Show HN: A faster way to set up Flowise and Qdrant for AI agent development","updated_at":"2025-02-24T14:45:06Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"zX41ZdbW"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Vector indexes, large server: MariaDB, Qdrant and pgvector"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://smalldatum.blogspot.com/2025/02/vector-indexes-large-server-dbpedia.html"}},"_tags":["story","author_zX41ZdbW","story_43066713"],"author":"zX41ZdbW","created_at":"2025-02-16T09:29:04Z","created_at_i":1739698144,"num_comments":0,"objectID":"43066713","points":3,"story_id":43066713,"title":"Vector indexes, large server: MariaDB, Qdrant and pgvector","updated_at":"2025-02-16T14:54:25Z","url":"http://smalldatum.blogspot.com/2025/02/vector-indexes-large-server-dbpedia.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"eko"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"LLM with Ollama and similarity search with Qdrant, vector database"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://vincent.composieux.fr/article/llm-with-ollama-and-similarity-search-with-qdrant-vector-database"}},"_tags":["story","author_eko","story_39661751"],"author":"eko","created_at":"2024-03-10T19:17:56Z","created_at_i":1710098276,"num_comments":0,"objectID":"39661751","points":3,"story_id":39661751,"title":"LLM with Ollama and similarity search with Qdrant, vector database","updated_at":"2024-09-20T16:38:52Z","url":"https://vincent.composieux.fr/article/llm-with-ollama-and-similarity-search-with-qdrant-vector-database"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"todsacerdoti"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Company Spotlight: Qdrant Vector Database"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://blog.replit.com/qdrant-vectordb"}},"_tags":["story","author_todsacerdoti","story_38954975"],"author":"todsacerdoti","created_at":"2024-01-11T17:01:19Z","created_at_i":1704992479,"num_comments":0,"objectID":"38954975","points":3,"story_id":38954975,"title":"Company Spotlight: Qdrant Vector Database","updated_at":"2024-09-20T16:05:35Z","url":"https://blog.replit.com/qdrant-vectordb"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"AkritiU"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"GPT-4 and Qdrant synergize, transforming poetry with enhanced coherence and depth. Visit my medium article to view the code implementation: https://medium.com/@akriti.upadhyay/how-to-augment-gpt-4-with-qdrant-to-elevate-its-poetry-composition-capabilities-acbb7379346f"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"How to Augment GPT-4 with Qdrant to Elevate Its Poetry Composition Capabilities"}},"_tags":["story","author_AkritiU","story_38640934","ask_hn"],"author":"AkritiU","created_at":"2023-12-14T13:11:05Z","created_at_i":1702559465,"num_comments":0,"objectID":"38640934","points":3,"story_id":38640934,"story_text":"GPT-4 and Qdrant synergize, transforming poetry with enhanced coherence and depth. Visit my medium article to view the code implementation: https://medium.com/@akriti.upadhyay/how-to-augment-gpt-4-with-qdrant-to-elevate-its-poetry-composition-capabilities-acbb7379346f","title":"How to Augment GPT-4 with Qdrant to Elevate Its Poetry Composition Capabilities","updated_at":"2024-09-20T15:51:57Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Vardhanam"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Building a Multidocument Chatbot Using Mistral 7B, Qdrant, and LangChain"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://blog.stackademic.com/building-a-multidocument-chatbot-using-mistral-7b-qdrant-and-langchain-1d9982186736?gi=ae0edaf4a4ff"}},"_tags":["story","author_Vardhanam","story_38611288"],"author":"Vardhanam","created_at":"2023-12-12T12:26:03Z","created_at_i":1702383963,"num_comments":0,"objectID":"38611288","points":3,"story_id":38611288,"title":"Building a Multidocument Chatbot Using Mistral 7B, Qdrant, and LangChain","updated_at":"2024-09-20T15:48:31Z","url":"https://blog.stackademic.com/building-a-multidocument-chatbot-using-mistral-7b-qdrant-and-langchain-1d9982186736?gi=ae0edaf4a4ff"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"yoquan"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Pgvector 0.4.0 Performance (Vs Qdrant)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://supabase.com/blog/pgvector-performance"}},"_tags":["story","author_yoquan","story_37348654"],"author":"yoquan","created_at":"2023-09-01T09:43:17Z","created_at_i":1693561397,"num_comments":0,"objectID":"37348654","points":3,"story_id":37348654,"title":"Pgvector 0.4.0 Performance (Vs Qdrant)","updated_at":"2024-09-20T15:01:37Z","url":"https://supabase.com/blog/pgvector-performance"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"williamstein"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Why Rust? (Qdrant Vector Database)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://qdrant.tech/articles/why-rust/#"}},"_tags":["story","author_williamstein","story_35960700"],"author":"williamstein","created_at":"2023-05-16T12:09:37Z","created_at_i":1684238977,"num_comments":0,"objectID":"35960700","points":3,"story_id":35960700,"title":"Why Rust? (Qdrant Vector Database)","updated_at":"2024-09-20T14:03:54Z","url":"https://qdrant.tech/articles/why-rust/#"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"andre-z"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant raises \u20ac2M to build open source vector search"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://www.ibbventures.de/en/news/qdrant-financing"}},"_tags":["story","author_andre-z","story_29911359"],"author":"andre-z","created_at":"2022-01-12T19:02:33Z","created_at_i":1642014153,"num_comments":0,"objectID":"29911359","points":3,"story_id":29911359,"title":"Qdrant raises \u20ac2M to build open source vector search","updated_at":"2024-09-20T10:12:07Z","url":"https://www.ibbventures.de/en/news/qdrant-financing"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tosh"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant vector db powers GROK and the RAG system in OpenAI"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://twitter.com/altryne/status/1721989500291989585"}},"_tags":["story","author_tosh","story_38188663"],"author":"tosh","children":[38188896,38189034],"created_at":"2023-11-08T10:36:34Z","created_at_i":1699439794,"num_comments":2,"objectID":"38188663","points":2,"story_id":38188663,"title":"Qdrant vector db powers GROK and the RAG system in OpenAI","updated_at":"2024-09-20T15:41:18Z","url":"https://twitter.com/altryne/status/1721989500291989585"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Alyka"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Integration of Qdrant ANN vector database back end with txtai"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/qdrant/qdrant-txtai"}},"_tags":["story","author_Alyka","story_33246082"],"author":"Alyka","children":[33246083],"created_at":"2022-10-18T12:17:04Z","created_at_i":1666095424,"num_comments":1,"objectID":"33246082","points":2,"story_id":33246082,"title":"Integration of Qdrant ANN vector database back end with txtai","updated_at":"2024-09-20T12:16:30Z","url":"https://github.com/qdrant/qdrant-txtai"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Alyka"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Qdrant vector search engine v0.9.0 update went live"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/qdrant/qdrant/releases/tag/v0.9.0"}},"_tags":["story","author_Alyka","story_32387452"],"author":"Alyka","children":[32387453],"created_at":"2022-08-08T16:17:23Z","created_at_i":1659975443,"num_comments":1,"objectID":"32387452","points":2,"story_id":32387452,"title":"Qdrant vector search engine v0.9.0 update went live","updated_at":"2024-09-20T11:47:03Z","url":"https://github.com/qdrant/qdrant/releases/tag/v0.9.0"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Alyka"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"V0.8.0 Qdrant vector search engine went live"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"https://github.com/qdrant/qdrant/releases/tag/v0.8.0"}},"_tags":["story","author_Alyka","story_31680199"],"author":"Alyka","children":[31680200],"created_at":"2022-06-09T12:12:07Z","created_at_i":1654776727,"num_comments":1,"objectID":"31680199","points":2,"story_id":31680199,"title":"V0.8.0 Qdrant vector search engine went live","updated_at":"2024-09-20T11:21:39Z","url":"https://github.com/qdrant/qdrant/releases/tag/v0.8.0"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"diegoglozano"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Hey HN!
I'm Diego. I've been extensively using Qdrant for hybrid search, but there's no standard way to track and manage schema migrations.
I've used alembic in the past for relational databases, so I decided to develop something similar for Qdrant: revector. You write declarative YAML migrations, commit them next to your code, and apply or roll them back with a single static binary.
Repo: https://github.com/diegoglozano/revector\nDocs: https://diegoglozano.github.io/revector/
Thanks for taking a look, happy to hear your feedback!"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["qdrant"],"value":"Show HN: Revector \u2013 Alembic-style schema migrations for Qdrant"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/diegoglozano/revector"}},"_tags":["story","author_diegoglozano","story_48818426","show_hn"],"author":"diegoglozano","created_at":"2026-07-07T14:32:09Z","created_at_i":1783434729,"num_comments":0,"objectID":"48818426","points":2,"story_id":48818426,"story_text":"Hey HN!
I'm Diego. I've been extensively using Qdrant for hybrid search, but there's no standard way to track and manage schema migrations.
I've used alembic in the past for relational databases, so I decided to develop something similar for Qdrant: revector. You write declarative YAML migrations, commit them next to your code, and apply or roll them back with a single static binary.
Repo: https://github.com/diegoglozano/revector\nDocs: https://diegoglozano.github.io/revector/
Thanks for taking a look, happy to hear your feedback!","title":"Show HN: Revector \u2013 Alembic-style schema migrations for Qdrant","updated_at":"2026-07-08T08:07:56Z","url":"https://github.com/diegoglozano/revector"}],"hitsPerPage":50,"nbHits":220,"nbPages":5,"page":0,"params":"query=Qdrant&tags=story&hitsPerPage=50&advancedSyntax=true&analyticsTags=backend","processingTimeMS":4,"processingTimingsMS":{"_request":{"roundTrip":16},"afterFetch":{"format":{"total":1},"merge":{"mergeLoop":{"prepareNextHit":2,"total":2},"total":2},"total":2},"fetch":{"total":1},"total":4},"query":"Qdrant","serverTimeMS":6}