This release includes 1 security fix for security teams reviewing exposed deployments.
Topics
+12 more
Affected surfaces
ReleasePort's take
Light signalQdrant v1.18.1 introduces io_uring support for quantized multi‑vector scorers and adds authorization checks before snapshot file uploads.
Why it matters: Io_uring enables higher I/O throughput for quantized scoring; mandatory auth on snapshot uploads prevents unauthorized data ingestion.
Summary
AI summaryUpdates https, Bug Fixes, and Change log across a mixed release.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Security | Medium |
Authorize request before accepting snapshot file upload. Authorize request before accepting snapshot file upload. Source: llm_adapter@2026-05-22 Confidence: low |
— |
| Feature | Medium |
Validate vector dimensions before WAL write for async upserts. Validate vector dimensions before WAL write for async upserts. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Feature | Medium |
Refactor quantized multi-vector scorers for io_uring support. Refactor quantized multi-vector scorers for io_uring support. Source: llm_adapter@2026-05-22 Confidence: low |
— |
| Performance | Medium |
Do not use unnecessary sequential mmaps. Do not use unnecessary sequential mmaps. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Performance | Medium |
Recreate workers/optimizers async to not block consensus. Recreate workers/optimizers async to not block consensus. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Notify pending consensus ops on snapshot apply. Notify pending consensus ops on snapshot apply. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Fix indexed integer range filter with float values. Fix indexed integer range filter with float values. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Fix `{match: {except: []}}` returning zero results with payload index. Fix `{match: {except: []}}` returning zero results with payload index. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Fix resharding cleanup data race with update queue. Fix resharding cleanup data race with update queue. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Restart optimizer on conflicting vector schema. Restart optimizer on conflicting vector schema. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Fix snapshot transfer cancellation on external malformed requests. Fix snapshot transfer cancellation on external malformed requests. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Fix deleted mask race on optimization and snapshot creation. Fix deleted mask race on optimization and snapshot creation. Source: llm_adapter@2026-05-22 Confidence: high |
— |
| Bugfix | Medium |
Fix empty vector panic. Fix empty vector panic. Source: llm_adapter@2026-05-22 Confidence: low |
— |
| Bugfix | Medium |
Fix TurboQuant heap memory under-reporting. Fix TurboQuant heap memory under-reporting. Source: llm_adapter@2026-05-22 Confidence: low |
— |
| Bugfix | Medium |
Use cluster default shard transfer method for fallback. Use cluster default shard transfer method for fallback. Source: llm_adapter@2026-05-22 Confidence: low |
— |
| Bugfix | Medium |
Do not remove shard for snapshot transfer receiver. Do not remove shard for snapshot transfer receiver. Source: llm_adapter@2026-05-22 Confidence: low |
— |
Full changelog
Change log
Improvements
- https://github.com/qdrant/qdrant/pull/8988 - Refactor quantized multi-vector scorers for io_uring support
- https://github.com/qdrant/qdrant/pull/9058 - Validate vector dimensions before WAL write for async upserts
Bug Fixes
- https://github.com/qdrant/qdrant/pull/8990 - Notify pending consensus ops on snapshot apply
- https://github.com/qdrant/qdrant/pull/9054 - Fix indexed integer range filter with float values
- https://github.com/qdrant/qdrant/pull/9055 - Fix
{match: {except: []}}returning zero results with payload index - https://github.com/qdrant/qdrant/pull/9070 - Fix empty vector panic
- https://github.com/qdrant/qdrant/pull/9099 - Fix TurboQuant heap memory under-reporting
- https://github.com/qdrant/qdrant/pull/9107 - Do not use unnecessary sequential mmaps
- https://github.com/qdrant/qdrant/pull/9014 - Fix resharding cleanup data race with update queue
- https://github.com/qdrant/qdrant/pull/9120 - Use cluster default shard transfer method for fallback
- https://github.com/qdrant/qdrant/pull/9121 - Recreate workers/optimizers async to not block consensus
- https://github.com/qdrant/qdrant/pull/9110 - Restart optimizer on conflicting vector schema
- https://github.com/qdrant/qdrant/pull/9122 - Do not remove shard for snapshot transfer receiver
- https://github.com/qdrant/qdrant/pull/9125 - Fix snapshot transfer cancellation on external malformed requests
- https://github.com/qdrant/qdrant/pull/9116 - Fix deleted mask race on optimization and snapshot creation
Security
- https://github.com/qdrant/qdrant/pull/9031 - Authorize request before we accept snapshot file upload
Security Fixes
- Authorize request before snapshot file upload — prevents unauthorized snapshot uploads (GHSA prefix not provided)
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About qdrant
Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
Beta — feedback welcome: [email protected]