Releasesby GlyphQdrant 1.18.1
Change log Improvements #8988 - Refactor quantized multi-vector scorers for io_uring support #9058 - Validate vector dimensions before WAL write for async upserts Bug Fixes #8990 - Notify pending consensus ops on snapshot apply #9054 - Fix indexed integer range filter with float values #9055 - Fix {match: {except: []}} returning zero results with payload...
Also shipped May 22, 2026Qdrant in May 2026 →
Prisma 8 Early Access: Write Your Contract, Prompt Your Agent, Ship Your App
Prisma 8 is now in Early Access, allowing developers to define their data layer as a contract. The framework handles migrations, type-safe queries, and continuous upgrades, designed to be compatible with AI coding agents.
What's new in Sigma
AI Apps Sigma Public Sigma Public is now available as a free-to-use platform where you can create, share, and explore apps built in Sigma. Use sample data or upload your own datasets to build workflows in an embedded Sigma experience. You can also search apps created by other users to get inspiration for your next app in Sigma.
Do more with grid
Grid is generally available and now makes it easier to reorder columns and rows, auto-position items into empty cells, and auto-add or remove rows to fit so you spend less time on manual fixes.
- Postgres wire protocol endpoint support for Looker: The Postgres Endpoint , a Postgres wire...
- Postgres wire protocol endpoint support for Looker: The Postgres Endpoint , a Postgres wire protocol-compatible MotherDuck client, now enables connectivity with Looker . DuckDB & MotherDuck plugin for Obsidian: Run DuckDB SQL or MotherDuck queries from inside an Obsidian note, with results frozen as a markdown table directly below the query.
Higher manual snapshot limit on paid plans, Neon MCP branch forking, neon-for-agent-platforms skill, and Neon branching video release - May 22, 2026
Higher manual snapshot limit on paid plans. Paid plans now include 100 manual snapshots per project, up from 10. Snapshots created by backup schedules do not co...
Railway Agent sandbox, HA static egress, GitHub org guardrails
We gave the Railway Agent a computer. The chat agent now gets a sandbox VM of its own, so it can clone a repo, run commands, edit files, and open a pull request without you copy-pasting code into the conversation. Static IP egress now supports high-availability with traffic distributed across three gateways and automatic failover.
More from QdrantEvery Qdrant release →
Qdrant 1.18.0
Change log Features 🏎️ milestone!48 - Add TurboQuant quantization variant, 8x vector compression without the recall tax #8605 - Add API to create and delete named vectors in existing collection #8606 - Deep memory reporting, show memory usage breakdown for storage components #8714 - Add low memory mode, force open everything on disk to minimize out of...
Qdrant 1.18.2
Change log Improvements #9282 - Log slow operations during shard WAL recovery #9137 - Clear cache of ID tracker after building a segment #9138 - Don't rebuild payload index if changing on_disk flag Bug Fixes #9285 - Fix potential infinite loop in optimizer when using multi vectors with prevent_unoptimized #9217 - Clean up unfinished segment optimizations on...
Qdrant 1.18.3
Change log Bug Fixes #9882 - Fix query errors when using shard keys while resharding
Qdrant 1.17.1
Change log Improvements milestone#46 - Defer point updates, efficiently apply and optimize points with prevent_unoptimized=true #8188 - Make Gridstore flushes non-blocking to reduce search tail latencies #8235 - Improve performance of filtered search in case of singular payload value #8402 - Add request tracing ID into audit log #8460 - Propagate WAL errors...
Qdrant 1.19.0
Change log Features ☝️ https://github.com/qdrant/qdrant/milestone/50 - TurboQuant 4-bit as a datatype of primary vector storage. Only store 4-bit quantized vectors and spare disk space on original vectors. [ docs ] #9669 , #9684 , #9950 - Unify definition of memory usage strategy for collection components.
Qdrant 1.17.0
Change log Features 🏋️ milestone#38 - Relevance Feedback ( docs ) milestone#44 - API for detailed report on optimization progress and stages ( docs ) milestone#40 - API for aggregated telemetry of the whole cluster ( docs ) milestone#43 - Unlimited update queue to gracefully smooth update spikes ( docs ) #8071 - Add Audit Access Logging ( docs ) #8063 -...
Qdrant 1.19.1
Change log Improvements #10342 , #10420 - Speed up and saturate memory bandwidth in quantized scoring with prefetching #10391 , #10392 , #10362 , #10437 , #10439 , #10381 - Rework and batch 4-bit TurboQuant SIMD implementation for better performance #10441 - Limit query plane allocations for better search performance #8650 - Improve L2 cosine preprocessing...
Qdrant 1.19.2
Change log Improvements - - Don't rewrite full sparse posting list on every upsert, making sparse vector upserts significantly faster - , , , - Make proxy segments persist their state, prevents WAL pinning and long recovery times on restart - - Fix HNSW builds not always saturating all assigned threads - - Resolve filter IDs only once per search request...