YOU JUST NEED POSTGRESStop managing 7 databases. Stop paying for 7 services. Stop waking up at 3 AM for 7 reasons. PostgreSQL can replace Redis, Elasticsearch, MongoDB, Kafka, and more — with the same algorithms.
It’s 2026, Just Use Postgres | Tiger Data
Think of your database like your home. Your home has a living room, bedroom, bathroom, kitchen, and garage. Each room serves a different purpose. But they're all under the same roof, connected by hallways and doors. You don't build a separate restaurant building just because you need to cook. You don't construct a commercial garage across town just to park your car. That's what Postgres is. One ho
So you wanna build a local RAG?
When we launched Skald, we wanted it to not only be self-hostable, but also for one to be able to run it without sending any data to third-parties. With LLMs getting better and better, privacy-sensitive organizations shouldn't have to choose between being left behind by not accessing frontier models and doing away with their committment to (or legal requirement for) data privacy. So here's what we
PGlite + pgvector ã§100è¡Œã§å®Ÿè£…ã™ã‚‹ãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢ (node/deno/drizzle)
pglite + pgvector ã§æ–‡ç« ã®é¡žä¼¼åº¦æ¤œç´¢ã‚’実装ã—ã¾ã™ã€‚ å‹•æ©Ÿ ã¨ã«ã‹ã手ã£å–ã‚Šæ—©ããƒãƒ¼ã‚«ãƒ«ã«ãƒ‡ãƒ¼ã‚¿ã‚’çªã£è¾¼ã‚“ã§ãŠã„ã¦æ¤œç´¢ã™ã‚‹ RAG ã®é››å½¢ãŒã»ã—ã‹ã£ãŸã‚“ã§ã™ãŒã€èª¿ã¹ã¦ã‚‚大è¦æ¨¡ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚’å‰æã¨ã—ãŸå¤§æŽ›ã‹ã‚Šãªå®Ÿè£…ãŒå¤šã„ã§ã™ã€‚ スクリプトを書ã„ãŸã‚‰ãƒãƒ³ã¨å®Ÿè¡Œã§ãるセットアップä¸è¦ãªã‚‚ã®ãŒã‚ã‚‹ã¨ã€è‰²ã€…ã¨å®Ÿé¨“ãŒã§ãã¾ã™ã€‚ mastra/rag ã‚’èªã‚“ã§ãŸã‚‰ã€ç°¡å˜ã«ã§ãã‚‹æ°—ãŒã—ãŸã®ã§ã‚„ã‚Šã¾ã—ãŸã€‚ãŸã ã€chunk ã®ãƒ‰ã‚ュメント分割相当ã®ã‚‚ã®ã¯ã¾ã 作ã£ã¦ã¾ã›ã‚“。ãã“ã¾ã§é›£ã—ã„概念ã§ã‚‚ãªã„ã®ã§ã€é›‘ã«ä½œã‚Œãã†ã§ã¯ã‚ã‚Šã¾ã™ã€‚ qrdrant も検討ã—ã¾ã—ãŸãŒã€ã‚µãƒ¼ãƒãƒ¼ã‚’建ã¦ã‚‹ã®ãŒé¢å€’ã§ã—㟠準備: ベクトル化用ã®é–¢æ•° 今回㯠@ai-sdk/openai を使ã£ã¦ãƒ™ã‚¯ãƒˆãƒ«åŒ–ã‚’ã—ã¾ã™ // OPENAI_API_KEY= import { openai } from "@ai-sdk/open
ベクトル検索エンジン Supabase with pgVector
ã“ã®è¨˜äº‹ã¯Supabaseã®ãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢æ©Ÿèƒ½ã®ç´¹ä»‹ã§ã™ã€‚ Supabase ベクトル検索エンジン Supabaseã¯åŸºæœ¬çš„ã«ã¯ã‚µãƒ¼ãƒãƒ¬ã‚¹ã®Postgresデータベースã§ã™ãŒã€pgvectorã¨ã„ã†æ‹¡å¼µã‚’使用ã™ã‚‹ã“ã¨ã§ãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢ã‚¨ãƒ³ã‚¸ãƒ³ã¨ã—ã¦ã‚‚使用ã™ã‚‹ã“ã¨ãŒã§ãã¾ã™ã€‚ ベクトル検索エンジンã¨ã¯ã€æ–‡ç« ・画åƒãƒ»éŸ³å£°ãªã©ã‚’ベクトル化ã—ã€ãƒ™ã‚¯ãƒˆãƒ«é–“ã®ã‚³ã‚µã‚¤ãƒ³é¡žä¼¼åº¦ã‚’使用ã—ã¦é¡žä¼¼æ€§ã®é«˜ã„コンテンツをå–å¾—ã™ã‚‹DBã§ã™ã€‚ ãã—ã¦ã€ãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢ã‚¨ãƒ³ã‚¸ãƒ³ã¨RDBを悪é”åˆä½“ã•ã›ã‚‹ã“ã¨ã§ã€SQLã®WHEREå¥ã‚„テーブル・ジョインãªã©RDBã®æ©Ÿèƒ½ã‚‚使ãˆã‚‹ã€ã„ã„ã¨ã“ã©ã‚ŠãªDBã«ãªã‚Šã¾ã™ã€‚
Amazon Aurora Serverless v2ã«pgvectorをインストールã—ã¦OpenAIã®embeddingãƒ‡ãƒ¼ã‚¿ã‚’æ ¼ç´ã™ã‚‹ - Qiita
ã¯ã˜ã‚ã« OpenAIã«å§‹ã¾ã‚Šã€AWSã§ã‚‚2023/9/29ã«Amazon BedrockãŒGAã—ãŸã‚Šã¨ã€2023å¹´ã¯LLM関連ã®ãƒ‹ãƒ¥ãƒ¼ã‚¹ãŒéžå¸¸ã«å¤šããªã£ã¦ã„る。ãã®ä¸ã§ã‚‚ã€Vector DBã¨çµ„ã¿åˆã‚ã›ã¦RAGを用ã„ã‚‹ã“ã¨ã§ãƒãƒ«ã‚·ãƒãƒ¼ã‚·ãƒ§ãƒ³ã®å›žé¿ã‚’ã™ã‚‹ã“ã¨ãŒã§ãるよã†ã«ãªã‚‹ãŸã‚ã€éžå¸¸ã«é‡è¦ãªæŠ€è¡“ã«ãªã£ã¦ãã¦ã„る。 AWSã§ã‚‚ã€2023/7/13よりAurora(PostgreSQL互æ›)ãŒpgvectorをサãƒãƒ¼ãƒˆã—ã¦ã€ã‚ˆã‚ŠVector DBãŒèº«è¿‘ãªã‚‚ã®ã«ãªã£ãŸã€‚ 今回ã¯ã€ã“ã®pgvectoræ‹¡å¼µã®ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«ã¨ä½¿ã„方を確èªã—ã¦ã„ã。 Amazon Aurora(PostgreSQL互æ›)を準備ã™ã‚‹ ã¾ãšã¯ã€Amazon Auroraクラスタを起動ã—よã†ã€‚ Amazon Auroraクラスタã®èµ·å‹•ã¯éŽåŽ»ã®è¨˜äº‹ã‚’å‚考ã«ã—ã¦ã„ãŸã ããŸã„。 ãªãŠã€pgvectorã®ã‚µãƒãƒ¼ãƒˆã—ã¦ã„ã‚‹Postgr
Amazon RDS for PostgreSQLãŒpgvectorモジュールã«å¯¾å¿œã—ベクトル検索ã§ãるよã†ã«ãªã‚Šã¾ã—㟠| DevelopersIO
è¿‘å¹´ã¯æ©Ÿæ¢°å¦ç¿’ã®ãƒ¢ãƒ‡ãƒ«ãƒ»ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ を使ã£ã¦ã‚ªãƒ–ジェクトをベクトル表ç¾ã—(embedding)ã€ãƒ†ã‚ストやメディアãªã©æ§˜ã€…ãªã‚ªãƒ–ジェクトã®æ¤œç´¢ã‚„分類ãŒç››ã‚“ã«è¡Œã‚ã‚Œã¦ãŠã‚Šã€LLMã®æµè¡Œã¨å…±ã«ãã®å‹¢ã„ãŒå¢—ã—ã¦ã„ã¾ã™ã€‚ RDBã®PostgreSQLå‘ã‘ã«ãƒ™ã‚¯ãƒˆãƒ«ç®¡ç†ãƒ»æ¤œç´¢ã‚’è¡Œã†æ‹¡å¼µãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ« pgvector ãŒAmazon RDS for RDSã§ã‚‚利用ã§ãるよã†ã«ãªã‚Šã€Pineconeã®ã‚ˆã†ãªãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢ç‰¹åŒ–åž‹ã®ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã‚’用æ„ã™ã‚‹ã“ã¨ãªãã€RDSインスタンスå˜ä½“ã§ãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢ãŒã§ãるよã†ã«ãªã‚Šã¾ã—ãŸã€‚ データベースã®ãƒžãƒãƒ¼ã‚¸ãƒ‰ã‚µãƒ¼ãƒ“スã¨ã—ã¦ã¯ã€ã“ã‚Œã¾ã§ã‚‚ Supabase ç‰ãŒ pgvector ã«å¯¾å¿œã—ã¦ã„ã¾ã—ãŸãŒã€RDS PostgreSQL ã‚‚ã¤ã„ã«å¯¾å¿œã—ã¾ã—ãŸã€‚ åˆ¶é™ pgvector 対応ã™ã‚‹æœ€è¿‘å‚探索 exact Approximation 対応è·é›¢ ユークリッド(L2)è·é›¢
1
メンテナンス
ランã‚ング
ãŠçŸ¥ã‚‰ã›
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Vercel Postgres AI-powered Semantic Search Demo
Introducing Vector Search with pgvector in CockroachDB
pgvectorを有効化ã—ãŸAurora Serverless v2ã®ã‚µãƒ¼ãƒãƒ¼ãƒ¬ã‚¹DBインスタンスã«LangChainã‹ã‚‰æŽ¥ç¶šã—ã¦ã¿ãŸ | DevelopersIO
Amazon Aurora PostgreSQL now supports pgvector for vector storage and similarity search - AWS
{{#tags}}- {{label}}
{{/tags}}