> ## Documentation Index
> Fetch the complete documentation index at: https://docs.firebolt.io/llms.txt
> Use this file to discover all available pages before exploring further.

> Latest release notes for the Firebolt data warehouse.

# Release notes

Firebolt continuously releases updates so that you can benefit from the latest and most stable service. These updates might happen daily, but we aggregate release notes to cover a longer time period for easier reference. The most recent release notes from the latest version are below.

* See the [Release notes archive](/release-notes/release-notes-archive) for earlier-version release notes.

<Note>
  **The rollout of version 4.32 is expected to be completed by the end of June 2026.**
  Firebolt might roll out releases in phases. New features and changes may not yet be available to all accounts.
</Note>

## Firebolt Release Notes - Version 4.32

### New Features

**Iceberg catalog databases with `CREATE ICEBERG DATABASE`**
Firebolt now supports creating databases backed by an Apache Iceberg catalog using [`CREATE ICEBERG DATABASE`](/reference-sql/commands/data-definition/create-iceberg-database). Both file-based catalogs and REST-shaped catalogs are supported, including AWS Glue, Snowflake Open Catalog, and Databricks. Tables in the catalog can be queried directly, and the available catalogs are listed in [`information_schema.catalogs`](/reference-sql/information-schema/catalogs). For REST-shaped catalogs, object-storage credentials can be supplied in a dedicated `STORAGE_CREDENTIALS` clause (separate from the catalog `CREDENTIALS` used for catalog authentication), which Firebolt uses as a fallback when the catalog does not vend its own storage credentials.

**Non-constant `LIKE` patterns**
[`LIKE`](/reference-sql/functions-reference/string/like), [`ILIKE`](/reference-sql/functions-reference/string/ilike), and [`regexp_like`](/reference-sql/functions-reference/string/regexp-like) now accept non-constant (dynamic) patterns, for example a pattern that comes from a column.

**Four-argument PostgreSQL-compatible `regexp_replace`**
Added support for the four-argument PostgreSQL form [`regexp_replace(string, pattern, replacement, flags)`](/reference-sql/functions-reference/string/regexp-replace). This improves compatibility with PostgreSQL-based BI clients that emit the flags argument.

### Behavior Changes

**`GRANT` and `REVOKE` restricted on built-in roles**
Built-in roles (`account_admin`, `system_admin`, `security_admin`, and `public`) are no longer modifiable through `GRANT`, `REVOKE`, or `ALTER DEFAULT PRIVILEGES`. The one exception is the `public` role: when the new `enable_grant_revoke_on_public_role` setting is enabled and the caller is `account_admin`, `GRANT` and `REVOKE` on `public` are permitted. The setting defaults to `false`.

### Performance Improvements

**Fast path for simple `VALUES` clauses**
Simple `VALUES` clauses are now planned through a dedicated fast path, reducing planning overhead.

**`CASE WHEN` simplified to `COALESCE`**
The optimizer now rewrites applicable `CASE WHEN` expressions into `COALESCE`, producing simpler and faster plans.

**Faster optimization of deeply nested `CASE WHEN`**
Logical optimization of deeply nested `CASE WHEN` expressions no longer grows exponentially, preventing long planning times on such queries.

**Improved cardinality estimation for `UNION`**
Cardinality estimation for union nodes has been improved, leading to better plan choices for queries involving unions.

### Bug Fixes

**Invalid UTF-8 handling in `LIKE` rewriting**
Fixed an invalid UTF-8 bug in the `LIKE` operator rewrite rule that could produce incorrect results for certain patterns.

**Cross-schema `CLONE` stored the schema name instead of its identifier**
Fixed an issue where cloning a table across schemas stored the schema name rather than the schema UUID as the `schema_id`, which could lead to incorrect schema resolution.

**Handling of renamed columns**
Changing the primary index to a renamed column now actually rewrites the data.