From 75fe4e8f957467e8d0c4fb1364e72f0f19bb8994 Mon Sep 17 00:00:00 2001
From: Pushkinist <4850452+Pushkinist@users.noreply.github.com>
Date: Thu, 18 Jun 2026 15:05:20 +0700
Subject: [PATCH] chore(release): 0.2.3

---
 CHANGELOG.md | 33 ++++++++++++++++++++++++++++++++-
 Cargo.lock   | 24 ++++++++++++------------
 Cargo.toml   |  2 +-
 3 files changed, 45 insertions(+), 14 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0d22e64..a3911ea 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -7,6 +7,36 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.2.3] - 2026-06-18
+
+Multi-model registry hardening. Two `--registry` serving bugs fixed: the
+multimodal encoder-output cache no longer leaks vision/audio features across
+models, and eager model preload now respects `--max-loaded-models`. No breaking
+changes.
+
+### Fixed
+
+- **Multimodal encoder-output cache cross-model leak.** In `--registry`
+  multi-model mode the vision/audio encoder cache was keyed on the
+  post-preprocess content hash only, so a cached image encoding produced for one
+  model (projected to its `hidden_size`) was returned to a different model for
+  the same image — a vision-feature shape mismatch (HTTP 503) when the hidden
+  sizes differed. The cache key now folds in a stable per-model signature, so
+  entries are never shared across models; same-model repeats still hit. (#132)
+- **Registry eager-preload ignored `--max-loaded-models`.** `rmlx serve
+  --registry` preloaded every model at startup even with a smaller resident cap,
+  paying the full load cost for models that were immediately evicted (a
+  ~5-minute boot for a 13-model registry). Preload is now bounded to at most
+  `--max-loaded-models` entries (the alphabetically-first ids, since the
+  registry is id-sorted); the rest load on demand. (#133)
+
+### Changed
+
+- `README.md` refreshed to 0.2.3 with an accurate "What works" summary, and
+  `docs/CLI.md` documents that the multimodal cache key now includes model
+  identity (no cross-model sharing) and that registry preload is bounded to the
+  resident cap.
+
 ## [0.2.2] - 2026-06-18
 
 Multimodal release. Whisper transcription works end to end (decode correctness
@@ -342,7 +372,8 @@ inference + conversion backend for Apple Silicon — no Python at runtime.
 - Speculative drafters validated against their verifiers: Qwen 3.6 MTP sidecar
   and the Gemma 4 assistant drafter.
 
-[Unreleased]: https://github.com/Pushkinist/rMLX/compare/v0.2.2...HEAD
+[Unreleased]: https://github.com/Pushkinist/rMLX/compare/v0.2.3...HEAD
+[0.2.3]: https://github.com/Pushkinist/rMLX/releases/tag/v0.2.3
 [0.2.2]: https://github.com/Pushkinist/rMLX/releases/tag/v0.2.2
 [0.2.1]: https://github.com/Pushkinist/rMLX/releases/tag/v0.2.1
 [0.2.0]: https://github.com/Pushkinist/rMLX/releases/tag/v0.2.0
diff --git a/Cargo.lock b/Cargo.lock
index 9615257..6377e33 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1921,7 +1921,7 @@ checksum = "dc897dd8d9e8bd1ed8cdad82b5966c3e0ecae09fb1907d58efaa013543185d0a"
 
 [[package]]
 name = "rmlx-audio"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "miniz_oxide",
  "rmlx-core",
@@ -1939,7 +1939,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-cli"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "anyhow",
  "chrono",
@@ -1974,7 +1974,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-core"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "chrono",
  "libc",
@@ -1988,7 +1988,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-kv-quant"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "rmlx-core",
  "rmlx-mlx",
@@ -1998,7 +1998,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-kv-ssd"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "rmlx-core",
  "rmlx-kv-quant",
@@ -2014,7 +2014,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-loader"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "memmap2",
  "rayon",
@@ -2030,7 +2030,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-metrics"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "csv",
  "regex-lite",
@@ -2049,7 +2049,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-mlx"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "bindgen",
  "rmlx-core",
@@ -2061,7 +2061,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-models"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "criterion",
  "image",
@@ -2088,7 +2088,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-quant"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "criterion",
  "rmlx-core",
@@ -2099,7 +2099,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-runtime"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "rmlx-core",
  "rmlx-mlx",
@@ -2109,7 +2109,7 @@ dependencies = [
 
 [[package]]
 name = "rmlx-server"
-version = "0.2.2"
+version = "0.2.3"
 dependencies = [
  "anyhow",
  "axum",
diff --git a/Cargo.toml b/Cargo.toml
index c4651f0..4339d82 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -16,7 +16,7 @@ members = [
 ]
 
 [workspace.package]
-version = "0.2.2"
+version = "0.2.3"
 edition = "2021"
 rust-version = "1.95"
 license = "MIT OR Apache-2.0"