SKaiNET-developers · michalharakal · Jun 6, 2026 · Jun 6, 2026 · Jun 6, 2026
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,6 +7,31 @@ version line is kept in lock-step with the underlying SKaiNET engine
 The format roughly follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.28.1] — 2026-06-06
+
+Version-aligned with **SKaiNET 0.28.1**. Skips 0.26.x / 0.27.x —
+SKaiNET-transformers tracked the engine internally across that window without a
+tagged release.
+
+### Changed
+
+- **`gradle/libs.versions.toml` `skainet` pin: 0.27.0 → 0.28.1.** Picks up the
+  completed Kotlin DSL → StableHLO → IREE export path. SKaiNET 0.28.0/0.28.1
+  closed the remaining DAG-DSL export bugs: shape-changing ops now declare their
+  inferred output type instead of echoing operand-0 — `reshape`/`matmul`/`concatenate`
+  ([SKaiNET #673](https://github.com/SKaiNET-developers/SKaiNET/issues/673)) and
+  `conv1d`/`gather`/`maxpool2d`/`avgpool2d`/`flatten`
+  ([SKaiNET #675](https://github.com/SKaiNET-developers/SKaiNET/issues/675)) — and
+  `reduce_window` is emitted in IREE's generic region form. A full gemma3 graph
+  traced through `GemmaMlirDumpTest` / `GemmaTraceTest` now lowers to StableHLO
+  that `iree-compile`s to a `vmfb`. No transformers-side API changes; existing
+  callers compile unchanged.
+
+### Verified
+
+- `:llm-inference:gemma:jvmTest` green against the published SKaiNET 0.28.1
+  (`GemmaMlirDumpTest` 1/1, `GemmaTraceTest` 1/1).
+
 ## [0.25.0] — 2026-05-25
 
 Version-aligned with **SKaiNET 0.25.0**. Skips 0.24.x — SKaiNET-transformers has

diff --git a/README.md b/README.md
@@ -97,23 +97,28 @@ Honest status — see the project-status note at the top of this README.
   before extending scope.
 - Verify each generative architecture end-to-end with smoke tests.
 - Wire the **StableHLO / native compilation path** for full transformer models.
+  As of 0.28.1 a full gemma3 graph exports to StableHLO and `iree-compile`s to a
+  `vmfb` (`GemmaMlirDumpTest`); next is running the compiled module and extending
+  the same path to the other families.
 
 ## Current release
 
-The current release is **0.25.0** — version-aligned with **SKaiNET 0.25.0**.
-Skips 0.24.x: the engine bumped 0.23.1 → 0.25.0 in the same release window
-without a tagged 0.24.x on either side. Brings the new
-[hybrid adaptive DSL with optional dtype constraints](https://github.com/SKaiNET-developers/SKaiNET/pull/616)
-(`DTypePolicy.Any | Require | Prefer | OneOf`) to every `*NetworkLoader`,
-turns the catalog BOM-only so every internal build now exercises
-`sk.ainet:skainet-bom` end-to-end, and locks three reference models
-(`@Tag("smoke-reference")`) for the smoke tier.
+The current release is **0.28.1** — version-aligned with **SKaiNET 0.28.1**.
+Skips 0.26.x / 0.27.x: SKaiNET-transformers tracked the engine internally across
+that window without a tagged release. The headline is that the engine's
+**Kotlin DSL → StableHLO → IREE export path is now complete** — a full gemma3
+graph traces and lowers to StableHLO that `iree-compile`s to a `vmfb`
+(`GemmaMlirDumpTest` / `GemmaTraceTest` are green against 0.28.1). SKaiNET
+0.28.0/0.28.1 fixed the remaining export bugs: result-type inference for
+`reshape`/`matmul`/`concatenate` ([#673](https://github.com/SKaiNET-developers/SKaiNET/issues/673))
+and `conv1d`/`gather`/pooling/`flatten` shapes plus the `reduce_window` emission
+form ([#675](https://github.com/SKaiNET-developers/SKaiNET/issues/675)).
 
 The recommended way to consume is via the BOM. It pins every published `skainet-transformers-*` artifact and re-exports the upstream `sk.ainet:skainet-bom`, so the engine-side `sk.ainet.core:skainet-*` artifacts get the matching version too — you only need to declare the BOM version in one place.
 
 ```kotlin
 dependencies {
-    implementation(platform("sk.ainet.transformers:skainet-transformers-bom:0.25.0"))
+    implementation(platform("sk.ainet.transformers:skainet-transformers-bom:0.28.1"))
 
     // Versions resolved from the BOM:
     implementation("sk.ainet.transformers:skainet-transformers-core")
@@ -190,6 +195,18 @@ try (KLlamaSession session = KLlamaJava.loadGGUF(modelPath, /* systemPrompt */ n
 
 See `llm-test/llm-test-java/src/test/java/.../KLlamaJavaToolCallingTest.java` for a runnable reference.
 
+## What's new in 0.28.1
+
+- **Engine pin `skainet 0.27.0 → 0.28.1`.** Picks up the completed Kotlin DSL →
+  StableHLO → IREE export path. Every shape-changing op now declares its inferred
+  output type (`reshape`/`matmul`/`concatenate`, [#673](https://github.com/SKaiNET-developers/SKaiNET/issues/673);
+  `conv1d`/`gather`/pooling/`flatten`, [#675](https://github.com/SKaiNET-developers/SKaiNET/issues/675)),
+  and `reduce_window` is emitted in IREE's generic region form — so a full gemma3
+  graph traced via `GemmaMlirDumpTest` lowers to StableHLO that `iree-compile`s to
+  a `vmfb`. No transformers-side API changes; existing callers compile unchanged.
+- Verified end-to-end: `:llm-inference:gemma:jvmTest` green against the published
+  0.28.1 (`GemmaMlirDumpTest`, `GemmaTraceTest` pass).
+
 ## What's new in 0.25.0
 
 - **`DTypePolicy` on every `*NetworkLoader.fromGguf` / `.fromSafeTensors`

diff --git a/docs/modules/ROOT/nav.adoc b/docs/modules/ROOT/nav.adoc
@@ -10,7 +10,6 @@
 
 .How-to Guides
 * xref:how-to/add-model.adoc[Add a New Model Architecture]
-* xref:how-to/add-compute-backend.adoc[Add a Compute Backend]
 * xref:how-to/add-tool.adoc[Add a Custom Tool]
 * xref:how-to/run-unified-cli.adoc[Use the Unified CLI]
 * xref:how-to/benchmarking.adoc[Run Benchmarks]

diff --git a/docs/modules/ROOT/pages/how-to/add-compute-backend.adoc b/docs/modules/ROOT/pages/how-to/add-compute-backend.adoc
diff --git a/docs/modules/ROOT/pages/tutorials/getting-started-java.adoc b/docs/modules/ROOT/pages/tutorials/getting-started-java.adoc
@@ -25,7 +25,7 @@ In your `build.gradle.kts`:
 [source,kotlin]
 ----
 dependencies {
-    implementation(platform("sk.ainet.transformers:skainet-transformers-bom:0.25.0"))
+    implementation(platform("sk.ainet.transformers:skainet-transformers-bom:0.28.1"))
 
     implementation("sk.ainet.transformers:skainet-transformers-runtime-kllama")
     implementation("sk.ainet.transformers:skainet-transformers-agent")
@@ -41,7 +41,7 @@ Or in Maven (Maven needs the `-jvm` classifier suffix on platform artifacts):
     <dependency>
       <groupId>sk.ainet.transformers</groupId>
       <artifactId>skainet-transformers-bom</artifactId>
-      <version>0.25.0</version>
+      <version>0.28.1</version>
       <type>pom</type>
       <scope>import</scope>
     </dependency>

diff --git a/docs/modules/ROOT/pages/tutorials/llama3-tool-calling.adoc b/docs/modules/ROOT/pages/tutorials/llama3-tool-calling.adoc
@@ -52,7 +52,7 @@ The pieces you need live in three modules:
 [source,kotlin]
 ----
 dependencies {
-    implementation(platform("sk.ainet.transformers:skainet-transformers-bom:0.25.0"))
+    implementation(platform("sk.ainet.transformers:skainet-transformers-bom:0.28.1"))
 
     implementation("sk.ainet.transformers:skainet-transformers-runtime-kllama")
     implementation("sk.ainet.transformers:skainet-transformers-agent")

diff --git a/gradle.properties b/gradle.properties
@@ -1,5 +1,5 @@
 GROUP=sk.ainet.transformers
-VERSION_NAME=0.26.0
+VERSION_NAME=0.28.1
 
 POM_DESCRIPTION=SKaiNET-transformers
 

diff --git a/gradle/libs.versions.toml b/gradle/libs.versions.toml
@@ -1,5 +1,5 @@
 [versions]
-skainet = "0.27.0"
+skainet = "0.28.1"
 agp = "9.2.1"
 jacksonDatabind = "2.22.0"
 jsonSchemaValidator = "3.0.3"