v2.6.24

## v2.6.24

Release date: September 16, 2026

| Milvus Version | Python SDK Version | Node.js SDK Version | Java SDK Version | Go SDK Version |
| -------------- | ------------------ | ------------------- | ---------------- | -------------- |
| 2.6.24         | 2.6.17             | 2.6.17              | 2.6.25           | 2.6.24         |

We are excited to announce the release of Milvus v2.6.24! This release focuses on resource efficiency — lower peak memory during segment loading, smarter force-merge compaction planning, and faster filtered search — along with important fixes for strong-consistency queries, replica channel balancing, and binlog import.

### Improvements

- Reworked force-merge compaction planning with multi-round knapsack packing so merged segments track the configured target size more closely and ID preallocation no longer over-reserves ([#52243](https://github.com/milvus-io/milvus/pull/52243))
- Refined segment loading resource control with batch-level memory estimation, bounded column-group temporary memory, and streaming V3 index loading, substantially lowering peak memory during segment load ([#52670](https://github.com/milvus-io/milvus/pull/52670), [#52787](https://github.com/milvus-io/milvus/pull/52787))
- Enabled delayed S3 file open by bumping milvus-storage, avoiding unnecessary object storage connections ([#52788](https://github.com/milvus-io/milvus/pull/52788))
- Reduced log noise during group chunk creation by emitting a single summary line per chunk instead of one per field ([#52900](https://github.com/milvus-io/milvus/pull/52900))
- Added optional per-request weights for RRF reranking in hybrid search, available through `FunctionScore`, legacy rank params, the RESTful API, and the Go client's `WithWeights` helper ([#52910](https://github.com/milvus-io/milvus/pull/52910))
- Improved filtered search performance by eliminating an atomic refcount hotspot in the scalar filter evaluation path that accounted for roughly 48% of leaf CPU time ([#53070](https://github.com/milvus-io/milvus/pull/53070))
- Added admission control that rejects RESTful DQL requests with HTTP 429 before body decoding when the proxy DQL queue is full, reducing proxy CPU usage under request floods ([#53112](https://github.com/milvus-io/milvus/pull/53112))
- Reduced write amplification during Tantivy-based text and NGRAM index building, lowering disk I/O and index build cost ([#53063](https://github.com/milvus-io/milvus/pull/53063))
- Added the configurable `queryNode.segcore.interimIndex.growingBuildThreadRate` parameter to allow multi-threaded interim index building on growing segments ([#53034](https://github.com/milvus-io/milvus/pull/53034))
- Added idempotent broadcast support so that retried bulk import requests resolve to the existing task instead of creating duplicate imports ([#53236](https://github.com/milvus-io/milvus/pull/53236))
- Reduced lock contention on the search path by caching per-chunk row counts during filter expression evaluation ([#53240](https://github.com/milvus-io/milvus/pull/53240), [#53248](https://github.com/milvus-io/milvus/pull/53248))
- Added a cluster-version gate that automatically enables write-before function materialization only after the whole cluster has been upgraded, keeping behavior consistent during rolling upgrades ([#53262](https://github.com/milvus-io/milvus/pull/53262))

### Bug fixes

- Fixed Woodpecker WAL ignoring the configured `woodpecker.meta.prefix` and writing metadata under the wrong etcd path, and added client-side append batching for higher small-batch concurrent write throughput ([#50201](https://github.com/milvus-io/milvus/pull/50201))
- Fixed a deadlock during the switch to the streaming service that could leave DML requests hanging ([#52293](https://github.com/milvus-io/milvus/pull/52293))
- Fixed SASL/SCRAM-SHA-256 authentication failures against Apache Kafka 4.x brokers by upgrading librdkafka to 2.6.1 ([#52666](https://github.com/milvus-io/milvus/pull/52666))
- Fixed flush hanging forever when `common.storage.useLoonFFI` is enabled and `minio.ssl.tlsMinVersion` is left at its default value ([#52731](https://github.com/milvus-io/milvus/pull/52731))
- Fixed an issue where all channels of a replica were loaded onto a single query node after scaling out, causing repeated out-of-memory kills and leaving the replica unserviceable ([#53090](https://github.com/milvus-io/milvus/pull/53090))
- Fixed strong-consistency queries failing with `channel tsafe stalled` (error code 505) under heavy upsert workloads by isolating online write execution from segment loading on QueryNode ([#53139](https://github.com/milvus-io/milvus/pull/53139))
- Fixed an issue where a frozen streaming node outside the primary resource group could be unfrozen and reassigned channels unexpectedly ([#53230](https://github.com/milvus-io/milvus/pull/53230))
- Fixed an issue where C++ log files were written to the `/tmp` directory and could fill up disk space ([#53292](https://github.com/milvus-io/milvus/pull/53292))
- Fixed binlog import failing when a nullable vector field had no binlog in the source segment ([#53357](https://github.com/milvus-io/milvus/pull/53357), [#53364](https://github.com/milvus-io/milvus/pull/53364))
- Fixed incorrect search results on embedding-list (vector array) fields when trailing empty lists were present, and upgraded Knowhere to v2.6.21 ([#53496](https://github.com/milvus-io/milvus/pull/53496))
- Bumped Woodpecker to v0.1.44, fixing a duplicate-symbol link failure that broke GPU builds on the 2.6 branch ([#53495](https://github.com/milvus-io/milvus/pull/53495))