Release 0.16.6 (#363)

Co-authored-by: Cursor Assistant <assistant@cursor.com>

Author: filipstrand

CHANGELOG.md

@@ -5,6 +5,27 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.16.6] - 2026-02-20
+
+### ✨ Improvements
+
+- **SeedVR2 7B support**: Add support for the SeedVR2 7B upscaler variant.
+- **Qwen-Image parity with diffusers**: Align Qwen-Image behavior more closely with the diffusers reference implementation.
+- **FIBO scheduler default**: Default FIBO `generate_image` to `flow_match_euler_discrete`.
+
+### 🧰 DX & Maintenance
+
+- **Repo tooling cleanup**: Remove unused Cursor command wrappers from the repository.
+- **SeedVR2 7B test coverage**: Add image test support for the new SeedVR2 7B path.
+
+### 👩‍💻 Contributors
+
+- **@ciaranbor**
+- **@icelaglace**
+- **@filipstrand**
+
+---
+
 ## [0.16.5] - 2026-02-17
 
 ### ✨ Improvements

pyproject.toml

@@ -14,7 +14,7 @@ source-exclude = [
 
 [project]
 name = "mflux"
-version = "0.16.5"
+version = "0.16.6"
 description = "MLX native implementations of state-of-the-art generative image models."
 readme = "README.md"
 keywords = ["flux", "ai", "ml", "transformers", "mlx", "huggingface", "apple-silicon", "diffusers", "qwen", "qwen-image", "seedvr2", "z-image"]

src/mflux/models/seedvr2/README.md

@@ -11,7 +11,6 @@ SeedVR2 is more recent and the preferred method for high-fidelity upscaling and
 
 ```sh
 mflux-upscale-seedvr2 \
-  --model seedvr2-7b \
   --image-path "input.png" \
   --resolution 2160 \
   --softness 0.5
@@ -35,7 +34,7 @@ image.save("input_upscaled.png")
 ```
 </details>
 
-This will upscale the image such that the shortest side is 2160 pixels while maintaining the aspect ratio. If `--model` is omitted, MFLUX defaults to `seedvr2-3b`.
+This will upscale the image such that the shortest side is 2160 pixels while maintaining the aspect ratio. If `--model` is omitted, MFLUX defaults to `seedvr2-3b`. Pass `--model seedvr2-7b` to use the 7B model. 
 
 Instead of specifying a target resolution, you can also use `--resolution 2x` or `--resolution 3x` to upscale by a factor of 2 or 3 respectively.