Apache Iceberg Release History

← Back to Iceberg cheatsheet

Apache Iceberg^™ Release history

Major releases since 0.7, latest first - significant features only

1.10.0 ? expected H2 2026 Forecast

Drawn from active proposals and public community discussion. Not a commitment - scope can shift, slip, or be cut between now and the actual release.

More spec v3 features promoted to GA: row lineage, default value handling polish
Geometry / geography types reach broader engine support
Variant type performance and shredding improvements; engine adoption
Materialized views (preview) - declarative refresh of derived tables
REST catalog: signed URLs / vended credentials standardisation across engines
Native encryption-at-rest spec refinements
Multi-table transactions via the catalog (Polaris / Nessie alignment)
Better compaction strategies for MoR with deletion vectors

1.9.0 2026-03 Latest

Spec v3 features: deletion vectors GA, default column values, richer types (geometry, geography)
Variant semi-structured type for nested JSON-like data
Iceberg REST catalog: vended credentials maturity
Compaction: better small-file consolidation under MoR

1.8.0 2025-09

Spec v3 preview - deletion vectors, default values, row lineage
Variant type for semi-structured (JSON-like) data with shredding
MERGE planning improvements; column-projection-aware deletes
Optimised partition statistics file format (Puffin)

1.7.0 2024-11

Iceberg REST catalog hardening and view spec finalisation
Scan planning offloaded server-side for REST catalogs
Spark 4.0 readiness and broader engine support

1.6.0 2024-07

Stored procedures for table maintenance stable
Faster manifest rewrite for large tables
Better support for engines other than Spark (Trino, Dremio, DuckDB)

1.5.0 2024-03

Initial design work on spec v3 (deletion vectors design)
Parquet bloom-filter writes for high-cardinality columns
Improved schema-evolution diagnostics

1.4.0 2023-09

REST catalog reaches production maturity; many engines adopt it
View spec stability (cross-engine view definitions)
Z-order compaction strategy

1.3.0 2023-05

View support in the Iceberg spec
Branching and tagging GA across engines
Position delete file maintenance procedures

1.2.0 2023-02

Storage credentials vending via the catalog
Position delete read-side optimisations
Partition statistics files (Puffin sidecar)

1.1.0 2022-12

Major engine ecosystem expansion (Snowflake/Dremio/DuckDB ingress maturing)
Spark 3.3 / Flink 1.16 support
Snapshot expiration and orphan-file removal performance

1.0.0 2022-09 Major

Stable spec v2 declared - format frozen for compatibility
Production-readiness signal for the broader ecosystem
Spark/Flink/Trino integrations promoted to GA

0.14.0 2022-07

Branching and tagging (preview) for write-audit-publish workflows
REST catalog spec introduced
Schema-evolution improvements: drop and add column at any position
Spark Structured Streaming sink stabilisation

0.13.0 2022-02

Optimised metadata reads via manifest summarisation
Parquet vectorised reads for Spark
Tighter Trino and Hive integrations

0.12.0 2021-09

Row-level deletes (position and equality) - merge-on-read support
Spec v2 finalised
MERGE INTO in Spark SQL

0.11.0 2021-04

Spark SQL extensions for Iceberg-specific DDL
Custom partition transforms (bucket, truncate, year/month/day/hour)
Flink connector stabilisation

0.10.0 2020-11

Hidden partitioning with column transforms
Time travel via snapshot IDs and timestamps
Initial Flink read/write support

0.9.0 2020-08

Per-file data file metrics (min/max bounds, null counts)
Partition evolution without rewriting historical data
Spark 3.0 support

0.8.0 2019-11

Schema evolution: add/drop/rename/reorder columns safely
Snapshot expiration and rollback APIs
Hive catalog integration

0.7.0 2018-12 First

First Apache release after donation from Netflix
Format spec v1 - immutable data files indexed by manifests and snapshots
Spark integration for reads and appends