From 0f6a769488c9f309b3be59b2729426fb3427109f Mon Sep 17 00:00:00 2001 From: Pushkinist <4850452+Pushkinist@users.noreply.github.com> Date: Thu, 18 Jun 2026 08:50:09 +0700 Subject: [PATCH] chore(release): 0.2.2 Bump workspace version 0.2.1 -> 0.2.2 and add the CHANGELOG section. 0.2.2 is a multimodal release: reliable Whisper transcription + a model-agnostic rmlx transcribe CLI (#119), Gemma 4 12B unified any-to-any image+audio input (#120), Gemma 4 native audio through the serve path (#122), the unified-vision bidirectional-attention colour fix (#127), the templated probe-smoke seed (#121), plus cosign release signing / CI SHA pinning (#116) and release-script fixes (#115, #118). --- CHANGELOG.md | 78 +++++++++++++++++++++++++++++++++++++++++++++++++++- Cargo.lock | 24 ++++++++-------- Cargo.toml | 2 +- 3 files changed, 90 insertions(+), 14 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 00d7f26..0d22e64 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -7,6 +7,81 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 ## [Unreleased] +## [0.2.2] - 2026-06-18 + +Multimodal release. Whisper transcription works end to end (decode correctness ++ long-form) behind a new model-agnostic `rmlx transcribe` CLI; the dense +Gemma 4 12B `gemma4_unified` any-to-any architecture is now supported for image +and audio input; the standard Gemma 4 family gains native audio input through +the serve path; and the unified vision color-fidelity bug is fixed. Plus +release-signing and CI-hardening housekeeping. No breaking changes. + +### Added + +- **`rmlx transcribe