Releasesby GlyphSemantic Search in Airbyte Context Layer
Semantic search is now available for Airbyte sources synced to BigQuery and Snowflake. This feature allows AI agents to query data based on meaning rather than keywords.
Vector search releases
304 Vector search releases from data and AI companies, newest first
Wednesday, November 4, 20264 releases
Embed 5 and Rerank 5 Are Now Available on OpenRouter
Cohere made Embed 5 (Pro and Fast) and Rerank 5 (Pro and Fast) models available on OpenRouter, supporting multimodal inputs and unified embedding spaces.
Twice the included embedding tokens through December 31
Pinecone is doubling the included monthly embedding tokens for Starter and Builder organizations through December 31, 2026.
Friday, October 30, 20262 releases
New tutorial: Discover Models with the Python SDK and REST API
A new tutorial for OCI Generative AI shows how to explore the model catalog with the Python SDK and REST API. Learn step by step how to authenticate with OCI IAM request signing, filter models by region, capability, serving mode, and lifecycle status, and inspect API compatibility and configuration parameters.
Thursday, October 29, 20262 releases
AlloyDB AI adds pre-registered Gemini and reranking models in extension 1.6.5
AlloyDB AI version 1.6.5 adds support for newer Gemini models and the semantic-ranker-512 model while setting gemini-3.5-flash-lite as default.
LanceDB v0.41.0
LanceDB v0.41.0 upgrades to Lance 14.0.2 and changes embedding configurations to require explicit application approval before restoring remote-code models or custom endpoints.
Wednesday, October 28, 20261 release
AlloyDB AI and vector search features are generally available
AlloyDB now provides general availability for four-level ScaNN vector indexes on PostgreSQL 17, supporting up to 10 billion rows. It also includes automatic index maintenance and streaming functionality for improved recall on filtered searches.
Tuesday, October 27, 20261 release
Released pg\_pinecone
The pg_pinecone extension is now available, enabling native vector search within PostgreSQL. It allows users to perform vector searches using the local engine or offload indexing to Pinecone for larger datasets.
Monday, October 26, 20261 release
Embedding
- Added the dashboard:set-filters embed event, which changes filter values on an embedded dashboard. Added the dashboard:set-controls embed event, which lets you update one or more controls (field switchers, timeframe switchers, multi-field pickers, top N limits) on an embedded dashboard without reloading it.
Friday, October 23, 20261 release
Retrieve Cortex Search (Public Preview)
The CORTEX_SEARCH_RETRIEVE table function is now in public preview, allowing users to execute low-latency search queries directly from SQL for RAG and chatbot applications.
Thursday, October 22, 20263 releases
Amazon OpenSearch Service introduces intelligent ingestion for logs and documents in Amazon S3
Amazon OpenSearch Serverless now supports intelligent ingestion, which automatically parses, chunks, and generates vector embeddings for logs and documents stored in Amazon S3.
Qdrant 1.20.0
Qdrant 1.20.0 adds BM25 scoring, TurboQuant vector datatypes, per-element payload indexes, and improved HNSW performance.
Wednesday, October 21, 20263 releases
BigQuery AI.CHUNK_DOCUMENT function is generally available
Google Cloud announced the general availability of AI.CHUNK_DOCUMENT in BigQuery, a table-valued function that splits unstructured documents into semantically coherent text chunks optimized for vector embeddings and RAG workflows.
Create embeddings for search and retrieval through AI Gateway
Turn text into vectors with OpenAI, Gemini, Qwen, Tencent, and Mistral embedding models, using your ngrok access key and credits.
Announcing Cohere's Rerank 5 Models
We're pleased to announce the release of Rerank 5 , Cohere's most advanced reranking models for enterprise retrieval: sharper search, with fewer tokens sent to your LLM.
Tuesday, October 20, 20262 releases
Released Pinecone Terraform Provider v5.1.0
Released v5.1.0 of the Terraform Provider for Pinecone. This version adds integrated embedding for document indexes, so Pinecone can embed a string field's text with a hosted model. It also adds on-demand read capacity for BYOC indexes. Indexes created with the top-level embed attribute are now called legacy integrated indexes.
TimescaleDB 2.31.0 (2026-10-20)
This release contains performance improvements and bug fixes since the 2.30.2 release. We recommend that you upgrade at the next available opportunity. Highlighted features in TimescaleDB v2.31.0 - Improved performance for UPDATE and DELETE queries on the columnstore with vectorized key processing for bulk decompression.
Friday, October 16, 20261 release
What's new in Sigma
AI Anthropic as an AI provider (GA) Add an Anthropic API key to use models from Anthropic with Sigma AI features. When you set up Anthropic as your external AI provider, you use Anthropic for the large language model (LLM) and must also configure a second AI provider that has an embeddings model. For more details, see Add Anthropic as an AI provider .
Thursday, October 15, 20262 releases
Lakebase Search supports Chinese, Japanese, and Korean full-text search
The lakebase_tokenizer extension now supports keyword search over Chinese, Japanese, and Korean (CJK) text. These languages don't separate words with spaces, so native PostgreSQL full-text search can't split them into useful search terms.
Transformers Release v5.20.0
Release v5.20.0 New Model additions Muse Spark Open Muse Spark Open is a multimodal mixture-of-experts model from Meta that takes text, images, audio, and video as input and generates text. It has 816B total parameters (42B active per token) and a context window of up to 1,048,576 tokens shared across modalities.
Wednesday, October 14, 20263 releases
Amazon S3 is now available on the Vercel Marketplace and v0 by Vercel
Amazon S3, including S3 Tables and S3 Vectors, is now available natively on the Vercel Marketplace and v0. Developers can provision and connect object, vector, and open table storage directly within Vercel.
Nemotron 3.5 Lightning now available on AI Gateway for free
Nemotron 3.5 Lightning from NVIDIA is now available for free on AI Gateway as nvidia/nemotron-3.5-lightning-free , served by NVIDIA. Nemotron 3.5 Lightning is an open-weight reasoning and chat model for long-running agents, sub-agent work, chatbots, and RAG systems.
Amazon S3 joins the Vercel Marketplace
You can now create and connect Amazon S3 resources through the Vercel Marketplace without leaving your Vercel project dashboard. The Amazon S3 Marketplace integration automatically adds connection details as environment variables and lets you manage resources alongside your application infrastructure, with links to AWS for deeper configuration.
Tuesday, October 13, 20262 releases
Select AI Support for Gemini Enterprise Agent Platform
Select AI supports Gemini Enterprise Agent Platform as an AI provider. You can configure a Select AI profile to use Gemini Enterprise Agent Platform models for Select AI actions such as natural language to SQL generation, chat, showsql, explainsql, embeddings, and other Select AI capabilities.
Supported PostgreSQL extensions on Lovable Cloud
Lovable Cloud databases support a fixed set of PostgreSQL extensions , such as pg_cron , pg_net , vector , and postgis . Lovable now refuses CREATE EXTENSION for any other extension, in the project chat, the SQL editor, and the Lovable API, and builds the feature with plain SQL instead. Extensions already installed in your database keep working.
Monday, October 12, 20264 releases
We closed 330 of your most-requested issues with Prisma ORM 8
Prisma ORM 8 resolves 330 open issues and introduces support for typed JSON, polymorphism, partial unique indexes, PostGIS, pgvector, and TypeScript migrations. This release focuses on improving developer experience and extensibility.
Announcing consumption-based pricing for short-term memory in Amazon Bedrock AgentCore
Announcing consumption-based pricing for short-term memory in Amazon Bedrock AgentCore Amazon Bedrock AgentCore memory now prices short-term memory (STM) based on the data your agents ingest, retrieve, and store, replacing the previous flat rate per event.
Released Pinecone Terraform Provider v5.0.0
Released v5.0.0 of the Terraform Provider for Pinecone. This version uses API version 2026-07 and adds the schema style for describing indexes, with the schema and deployment attributes. Use it to manage document indexes with dense vector, sparse vector, and full-text-search fields.
Embedding
Added an outcome field to the embedded dashboard status event , so hosts can tell whether a dashboard reached done because it completed, a user canceled its queries, or a tile returned an error.
Friday, October 9, 20262 releases
Docling 2.136.0
Feature heading_hierarchy: Recognize Division/Subdivision heading markers in hierarchy inference (8a6155b) mlx: Add repetition penalty options (dd7c035) Fix ocr: Do not OCR programmatic text because a vector shape crosses it (c6ef3c4) latex: Resolve \input and \includegraphics for str sources (b2a59b9) latex: Keep CMYK JPEG figures (9943158) pptx: Move...
Llama Stack 1.5.0
Minor release off the new release-1.5.x branch. 108 commits since v1.4.0. Breaking changes Serve conversations and prompts with responses, or when explicitly listed An explicit apis: list is now authoritative, even when empty Raise the FastAPI floor to 0.137.2 and drop legacy route collection Allow an empty choices list in the OpenAI legacy completions...
Wednesday, October 7, 20262 releases
Weaviate 1.40.0 - Alter Schema Drop Vector Index, HFresh MUVERA, RQ-4 quantization
Breaking Changes none Namespaces (GA) Namespaces add control-plane and data isolation between users on a shared cluster. - Namespaces: Gate tenant actions - Namespaces: Add plumbing for namespace exister - Pin replacement replicas to a namespace's home node - Namespaces: Effective-suspended derivation + shard-materialization guards - Namespaces: Add support...
LanceDB v0.40.0
This stable release uses Lance 13.0.0 and is maintained on release/v0.40. Java packaging: Use com.lancedb:lancedb-core:0.40.0. The Maven artifacts were built from release-branch commit 711deef7, which corrects version serialization in the tag's Maven POMs. The Java runtime code is unchanged.
Tuesday, October 6, 20263 releases
PostgreSQL Advanced Edition clusters now support 20 additional extensions beyond the pre-installed pg_repack...
PostgreSQL Advanced Edition clusters now support 20 additional extensions beyond the pre-installed pg_repack, pg_stat_statements, pgaudit, plpgsql, and vector extensions, including pg_trgm, pgcrypto, citext, and uuid-ossp. You can enable …
Promptfoo 0.124.0
0.124.0 (2026-10-06) ⚠ BREAKING CHANGES providers: remove hosted ChatKit provider providers: make WatsonX SDKs opt-in deps: make Langfuse SDK opt-in deps: make Slack SDK opt-in providers: make OpenAI Agents an explicit install providers: make Codex Security SDK opt-in providers: make browser SDKs explicit installs deps: make local Transformers opt-in...
Transformers Release v5.19.0
Release v5.19.0 New Model additions EmbeddingGemma2 EmbeddingGemma 2 is a multimodal embedding model from Google built on the Gemma 4 architecture. It encodes text, images, audio, and video, individually or combined in one input, into a shared 768-dimensional vector space for cross-modal retrieval, semantic similarity, clustering, and classification.
Monday, October 5, 20261 release
The gemini-embedding-2 model works well for embedding long strings, including multilingual and unstructured...
The gemini-embedding-2 model works well for embedding long strings, including multilingual and unstructured data. It supports a mix of text, images, audio, video, and PDF files. You can use this model in the AI.EMBED , AI.SIMILARITY , and AI.GENERATE_EMBEDDING functions. This model is generally available .
Thursday, October 1, 20263 releases
AI Search is generally available
AI Search is now generally available. Usage-based billing begins on November 1, 2026, with included monthly ingestion, storage, semantic query, and full-text query usage. Cloudflare will send a reminder email the week before billing begins. Refer to Limits & pricing for rates and included usage.
- DuckDB 1.5.6 support: MotherDuck supports DuckDB 1.5.6
- DuckDB 1.5.6 support: MotherDuck supports DuckDB 1.5.6 . Refer to the changelog and the official announcement for details. Note: Our install scripts ensure you always have a MotherDuck-supported DuckDB client version.
Wednesday, September 30, 20267 releases
Transformers Release 5.18.0
New Model additions Nemotron 3 Diarization Nemotron 3 Diarization is an open-weight streaming speaker diarization model designed to determine "who spoke when" in real-world audio. It supports both streaming and offline inference, handles up to eight speakers, and orders speaker outputs by each speaker's first arrival in the input audio.
Voyage Rerank 3 and Rerank 3 Lite are now available on AI Gateway
Rerank 3 and Rerank 3 Lite from Voyage AI by MongoDB are now available on AI Gateway. The models reorder search results by relevance before passing documents to a language model. The series improves retrieval quality across domains, with the largest gains on long documents and code.
Amazon S3 Vectors introduces metadata pre-filtering for up to 5x higher recall on filtered search
Amazon S3 Vectors now supports pre-filtering, which evaluates metadata filters before running similarity search, returning up to 5x more of the matching vectors when your filter is selective. S3 Vectors also adds a prefix match operator ($startsWith) for filtering on values like paths and URLs.
You can no longer install extensions on PostgreSQL Advanced Edition clusters with CREATE EXTENSION
You can no longer install extensions on PostgreSQL Advanced Edition clusters with CREATE EXTENSION. The pre-installed pg_repack, pg_stat_statements, pgaudit, plpgsql, and vector extensions remain available. To request another extension, …
Announcing Cohere's Embed 5 Models
We're pleased to announce the release of Embed 5 , Cohere's most powerful embeddings family yet. Embed 5 delivers frontier retrieval quality on complex enterprise data, with major gains over Embed 4 on visually rich documents, financial filings, parsed PDFs, code, and multilingual retrieval.
Monday, September 28, 20264 releases
[Launched] Generally Available: Vector search and vector indexes in Azure SQL
Vector search and vector indexes are now generally available in Azure SQL Database. Using the CREATE VECTOR INDEX and VECTOR_SEARCH T-SQL capabilities, you can perform efficient approximate nearest neighbor (ANN) searches for AI scenarios such as semantic
Released Go SDK v7.0.0
Released v7.0.0 of the Pinecone Go SDK. This version uses the latest stable API version, 2026-07, and adds support for schema-based document indexes and the Documents API, which brings full-text search to the Go SDK. The v6 release line continues to target 2026-04.
Embedding
Added page:ready and dashboard:ready events that fire when embedded content has finished loading, and messages sent while the page loads are now held and delivered after the ready event fires.
Lakebase Search adds a configurable text tokenizer for keyword search
The new lakebase_tokenizer extension lets you control how Lakebase Search turns text into search terms, so keyword search matches the way your data and users actually read. Native PostgreSQL full-text search gives you little control over this step.
Friday, September 25, 20261 release
Ollama 0.40.0
Models run on MLX on Apple Silicon by default In this release, on Apple Silicon devices, model architectures supported by the MLX runtime will automatically run on MLX.
Thursday, September 24, 20262 releases
ai_search can search Lakebase synced tables (Beta)
The ai_search function can now retrieve from Lakebase synced tables with Lakebase Search enabled, in addition to AI Search indexes. Add a lakebase_search knowledge source to search a Lakebase synced table, or combine it with vector_search sources in the same call. This feature is in Beta . See ai_search function .
spark-nlp 7.0.0
📢 Spark NLP 7.0.0: Welcome, Apache Spark 4! New support for Spark 4.x alongside Spark 3.x Spark NLP is taking a major step forward with 7.0.0: support for Apache Spark 4.x and Scala 2.13, while preserving the Spark 3.x / Scala 2.12 line. This is more than a version-number update.
Wednesday, September 23, 20263 releases
Find models by region with model discovery in OCI Generative AI
OCI Generative AI now supports model discovery. Use model discovery to find models in a specified region and review details such as: Model capability including chat, text embedding, and text reranking. Input and output types, such as IMAGE_TEXT_TO_TEXT. Whether a model supports on-demand or dedicated serving.
Amazon Bedrock Managed Knowledge Base now supports Salesforce and Zendesk as native data source connectors
AWS announces Salesforce and Zendesk data source connectors for Amazon Bedrock Managed Knowledge Base, a fully managed retrieval-augmented generation (RAG) service. Customers can now sync Salesforce knowledge articles and Zendesk articles and community posts directly into their managed knowledge base.
Mem0 Python SDK (v2.2.0)
New Features: Client: Add User Profiles to MemoryClient and AsyncMemoryClient : get_profile() , generate_profile() , get_profile_settings() , update_profile_settings() , sample_profiles() , and get_profile_job() .
Tuesday, September 22, 20261 release
Select AI for Java
Select AI for Java provides a Java SDK for using Select AI capabilities in Java applications. Use the Java-native API to work with AI profiles, credentials, conversations, vector indexes, natural-language SQL, RAG, summarization, translation, feedback, and synthetic data generation. See Select AI for Java for more information.
Monday, September 21, 20262 releases
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes. This feature is generally available ( GA ). It accelerates vector search performance and increases queries per second (QPS) for vector workloads. For more information, see Accelerate queries with the Columnar Engine .
arize-phoenix-evals: v3.9.0
3.9.0 (2026-09-21) Features evals: add retrieval relevance evaluator (7efa7b2) evals: add session-level pii_detection evaluator (0e3b2c2) evals: completeness evaluator (63805d0) evals: deprecate document relevance evaluators (d67ea3f) Bug Fixes evals: accept role aliases in PromptTemplate message lists (767847c) evals: use non-blocking sleep in async rate...
Saturday, September 19, 20261 release
Temporal 1.32.0
Standalone Activities GA Standalone Activities and start delay are now enabled by default via activity.enableStandalone and activity.startDelayEnabled . This release adds delayed starts, operator APIs, and batch operations.
Friday, September 18, 20262 releases
What's new in Sigma
AI Migrate to Sigma with an AI assistant (Beta) You can use migration skills for Sigma to have an AI assistant rebuild your source content , the dashboards, reports, and data models in another business intelligence (BI) tool , as a Sigma document.
Lakebase Search is generally available
Lakebase Search is now generally available. It brings scalable vector and keyword search to Lakebase projects through two Postgres extensions, so you can run semantic and exact-term search in a single database without a separate search stack: lakebase_vector adds the lakebase_ann index type for approximate nearest-neighbor vector search.