diff --git a/CHANGELOG.md b/CHANGELOG.md
index 40af1ebcd..8c3dfe9ac 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -9,6 +9,8 @@ from version 5.0.0 onward. Pre-fork releases (`1.x`–`4.2.0`) were authored by
 
 ## [Unreleased]
 
+## [5.1.0] - 2026-08-29
+
 > The entries below also cover the **b9917 → b10456** window (PRs #341–#394), which went unrecorded
 > here while it happened; they were reconstructed from the git history and from
 > [`docs/history/llama-cpp-breaking-changes.md`](docs/history/llama-cpp-breaking-changes.md), which
@@ -520,7 +522,8 @@ Releases `1.1.1` through `4.2.0` were authored by [@kherud](https://github.com/k
 
 For an architecture-level diff between the pre-fork baseline (`49be664`) and the first 5.0.0 candidate (`24918e4`), see [`docs/history/49be664_24918e4.md`](docs/history/49be664_24918e4.md). For the server-fork-deletion refactor that culminated in 5.0.0, see [`docs/history/REFACTORING.md`](docs/history/REFACTORING.md). For the chat-completion integration that landed in 5.0.0, see [`docs/history/CHAT_INTEGRATION_SUMMARY.md`](docs/history/CHAT_INTEGRATION_SUMMARY.md).
 
-[Unreleased]: https://github.com/bernardladenthin/java-llama.cpp/compare/v5.0.6...HEAD
+[Unreleased]: https://github.com/bernardladenthin/java-llama.cpp/compare/v5.1.0...HEAD
+[5.1.0]: https://github.com/bernardladenthin/java-llama.cpp/compare/v5.0.6...v5.1.0
 [5.0.6]: https://github.com/bernardladenthin/java-llama.cpp/compare/v5.0.5...v5.0.6
 [5.0.5]: https://github.com/bernardladenthin/java-llama.cpp/compare/v5.0.4...v5.0.5
 [5.0.4]: https://github.com/bernardladenthin/java-llama.cpp/compare/v5.0.3...v5.0.4
diff --git a/README.md b/README.md
index 63d98f4a5..7cb85c66c 100644
--- a/README.md
+++ b/README.md
@@ -128,7 +128,7 @@ Access this library via Maven (released versions on Maven Central):
 <dependency>
     <groupId>net.ladenthin</groupId>
     <artifactId>llama</artifactId>
-    <version>5.0.6</version>
+    <version>5.1.0</version>
 </dependency>
 ```
 
@@ -218,7 +218,7 @@ there. Pick **at most one** classifier (they are mutually exclusive):
 <dependency>
     <groupId>net.ladenthin</groupId>
     <artifactId>llama</artifactId>
-    <version>5.0.6</version>
+    <version>5.1.0</version>
 </dependency>
 
 <!-- GPU / accelerator or alternate-CPU build: add the <classifier> from the
@@ -226,7 +226,7 @@ there. Pick **at most one** classifier (they are mutually exclusive):
 <dependency>
     <groupId>net.ladenthin</groupId>
     <artifactId>llama</artifactId>
-    <version>5.0.6</version>
+    <version>5.1.0</version>
     <classifier>cuda13-linux-x86-64</classifier>
 </dependency>
 ```
@@ -994,7 +994,7 @@ forcing that floor on every core consumer. It ships and versions in lockstep wit
 <dependency>
     <groupId>net.ladenthin</groupId>
     <artifactId>llama-langchain4j</artifactId>
-    <version>5.0.6</version>
+    <version>5.1.0</version>
 </dependency>
 ```
 
@@ -1154,12 +1154,12 @@ One dependency line in Android Studio — no submodule, no NDK build, no manual
 
 ```kotlin
 dependencies {
-    implementation("net.ladenthin:llama-android:5.0.6")
+    implementation("net.ladenthin:llama-android:5.1.0")
     // or, for Qualcomm Adreno GPUs (device must provide an OpenCL ICD):
-    // implementation("net.ladenthin:llama-android-opencl:5.0.6")
+    // implementation("net.ladenthin:llama-android-opencl:5.1.0")
 
     // optional Kotlin coroutines facade (Flow streaming + suspend wrappers):
-    implementation("net.ladenthin:llama-kotlin:5.0.6")
+    implementation("net.ladenthin:llama-kotlin:5.1.0")
 }
 ```
 
diff --git a/llama-android/README.md b/llama-android/README.md
index 2a8743aa7..7012b378b 100644
--- a/llama-android/README.md
+++ b/llama-android/README.md
@@ -13,9 +13,9 @@ ProGuard rules.
 ```kotlin
 // build.gradle.kts of your app — that's all.
 dependencies {
-    implementation("net.ladenthin:llama-android:5.0.6")
+    implementation("net.ladenthin:llama-android:5.1.0")
     // or, for Qualcomm Adreno GPUs (device must provide an OpenCL ICD):
-    // implementation("net.ladenthin:llama-android-opencl:5.0.6")
+    // implementation("net.ladenthin:llama-android-opencl:5.1.0")
 }
 ```
 
diff --git a/llama-kotlin/README.md b/llama-kotlin/README.md
index 3c30145ed..045b9a024 100644
--- a/llama-kotlin/README.md
+++ b/llama-kotlin/README.md
@@ -11,11 +11,11 @@ Pure Kotlin/JVM — works on desktop JVMs **and** Android.
 
 ```kotlin
 dependencies {
-    implementation("net.ladenthin:llama-kotlin:5.0.6")
+    implementation("net.ladenthin:llama-kotlin:5.1.0")
     // ...plus the binding itself — the façade does NOT drag it in transitively
     // (provided scope), so YOU pick the right flavor:
-    implementation("net.ladenthin:llama:5.0.6")          // desktop JVM
-    // implementation("net.ladenthin:llama-android:5.0.6") // Android (AAR)
+    implementation("net.ladenthin:llama:5.1.0")          // desktop JVM
+    // implementation("net.ladenthin:llama-android:5.1.0") // Android (AAR)
 }
 ```
 
diff --git a/llama-kotlin/pom.xml b/llama-kotlin/pom.xml
index 15f553bb2..01bd93981 100644
--- a/llama-kotlin/pom.xml
+++ b/llama-kotlin/pom.xml
@@ -12,7 +12,7 @@ SPDX-License-Identifier: MIT
 	<parent>
 		<groupId>net.ladenthin</groupId>
 		<artifactId>llama-parent</artifactId>
-		<version>5.0.7-SNAPSHOT</version>
+		<version>5.1.0</version>
 		<relativePath>../pom.xml</relativePath>
 	</parent>
 
diff --git a/llama-langchain4j/README.md b/llama-langchain4j/README.md
index 8628c334a..314039ebc 100644
--- a/llama-langchain4j/README.md
+++ b/llama-langchain4j/README.md
@@ -62,7 +62,7 @@ ScoringModel reranker     = new JllamaScoringModel(rerankLlama);
 <dependency>
     <groupId>net.ladenthin</groupId>
     <artifactId>llama-langchain4j</artifactId>
-    <version>5.0.6</version>
+    <version>5.1.0</version>
 </dependency>
 ```
 
diff --git a/llama-langchain4j/pom.xml b/llama-langchain4j/pom.xml
index 3793b60b5..5f6558528 100644
--- a/llama-langchain4j/pom.xml
+++ b/llama-langchain4j/pom.xml
@@ -12,7 +12,7 @@ SPDX-License-Identifier: MIT
 	<parent>
 		<groupId>net.ladenthin</groupId>
 		<artifactId>llama-parent</artifactId>
-		<version>5.0.7-SNAPSHOT</version>
+		<version>5.1.0</version>
 		<relativePath>../pom.xml</relativePath>
 	</parent>
 
diff --git a/llama/pom.xml b/llama/pom.xml
index 2134d3e6a..351c28511 100644
--- a/llama/pom.xml
+++ b/llama/pom.xml
@@ -13,7 +13,7 @@ SPDX-License-Identifier: MIT
 	<parent>
 		<groupId>net.ladenthin</groupId>
 		<artifactId>llama-parent</artifactId>
-		<version>5.0.7-SNAPSHOT</version>
+		<version>5.1.0</version>
 		<relativePath>../pom.xml</relativePath>
 	</parent>
 
@@ -93,7 +93,7 @@ SPDX-License-Identifier: MIT
 		<spotless.version>3.10.1</spotless.version>
 		<palantir-java-format.version>2.97.0</palantir-java-format.version>
 		<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
-		<project.build.outputTimestamp>2026-07-07T07:32:17Z</project.build.outputTimestamp>
+		<project.build.outputTimestamp>2026-08-29T16:06:02Z</project.build.outputTimestamp>
 	</properties>
 
 	<!--
diff --git a/pom.xml b/pom.xml
index dd1250562..c8e14c6f3 100644
--- a/pom.xml
+++ b/pom.xml
@@ -19,7 +19,7 @@ SPDX-License-Identifier: MIT
 	-->
 	<groupId>net.ladenthin</groupId>
 	<artifactId>llama-parent</artifactId>
-	<version>5.0.7-SNAPSHOT</version>
+	<version>5.1.0</version>
 	<packaging>pom</packaging>
 
 	<name>${project.groupId}:${project.artifactId}</name>