diff --git a/CITATION.cff b/CITATION.cff
index 50fc2b83..47f321d3 100644
--- a/CITATION.cff
+++ b/CITATION.cff
@@ -9,4 +9,4 @@ authors:
 repository-code: "https://github.com/derek73/python-nameparser"
 url: "https://github.com/derek73/python-nameparser"
 license: LGPL-2.1-or-later
-version: "2.2.0"
+version: "2.3.0"
diff --git a/docs/release_log.rst b/docs/release_log.rst
index c66a2e9f..e15e2f1a 100644
--- a/docs/release_log.rst
+++ b/docs/release_log.rst
@@ -1,8 +1,23 @@
 Release Log
 ===========
-* 2.3.0 - Unreleased
-
-    nameparser 2.3 is under development.
+* 2.3.0 - September 12, 2026
+
+    nameparser 2.3 is parsing fixes and new honorific vocabulary;
+    nothing in the API is removed or renamed.
+
+    The fixes cluster around post-nominals and titles. A space-separated
+    run of post-nominals keeps the spacing the writer typed, and the
+    acronyms that are also surnames -- Rai, Cha, Ba -- no longer take a
+    name's family name. A run of titles addresses by its last, and a
+    trailing abbreviated title reads as a title. CJK names and honorifics
+    written with a full stop of any width now parse. The additions are
+    renunciate and royal given-name titles, Devanagari and the first
+    Bengali honorifics, and two ``AmbiguityKind`` members that report a
+    reading nothing in the name decided.
+
+    One incompatibility: a ``Lexicon`` pickled by 2.1.x or 2.2.x with a
+    caller-added wide-stop or NFD entry no longer loads; the full-stop
+    bullet below has the remedy.
 
     **Behavior Changes**
 
diff --git a/nameparser/_version.py b/nameparser/_version.py
index a607b7e4..d7df1fcc 100644
--- a/nameparser/_version.py
+++ b/nameparser/_version.py
@@ -15,5 +15,5 @@
 #: "dev" through the cycle, cleared at release. It normalizes to
 #: 2.3.0.dev0, which sorts above 2.2.0 and BELOW 2.3.0, so an install
 #: from master can never masquerade as the release it precedes.
-PRE_RELEASE = "dev"
+PRE_RELEASE = ""
 __version__ = ".".join(map(str, VERSION)) + PRE_RELEASE