You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: CHANGELOG.md
-112Lines changed: 0 additions & 112 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -5,118 +5,6 @@ All notable changes to Stability Matrix will be documented in this file.
5
5
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
6
6
and this project adheres to [Semantic Versioning 2.0](https://semver.org/spec/v2.0.0.html).
7
7
8
-
<<<<<<< HEAD
9
-
=======
10
-
## v2.16.0-dev.3
11
-
### Added
12
-
- Added enable/disable toggle for environment variables in Settings, allowing variables to be temporarily disabled without deleting them
13
-
- Added single-instance window activation signaling so reopening the app restores and focuses the existing desktop window instead of launching a duplicate instance
14
-
- Added notification system with localizable banner and markdown detail dialog UI
15
-
- Added warning in data directory selector when a OneDrive folder is selected
16
-
- Added support in the Checkpoints page to distinguish standard updates from Early Access-only updates - thanks to @x0x0b!
17
-
- Added torch index for Strix/Gorgon Point Ryzen AI APUs on Windows - thanks to @NeuralFault!
18
-
- Added retry button to failed downloads - thanks to @NeuralFault!
19
-
### Changed
20
-
- Improved safetensor checkpoint classification to correctly detect UNet-only models for Wan Video, HiDream, Z-Image, Hunyuan3D, and diffusers-format Flux architectures, ensuring they are routed to the DiffusionModels folder
21
-
- GGUF checkpoint downloads now go directly to the DiffusionModels folder instead of StableDiffusion
22
-
- Configured portable Git to suppress detached HEAD advice messages
23
-
- Settings file saves are now atomic to prevent corruption from interrupted writes
24
-
- Updated torch indexes for A1111, ComfyUI, InvokeAI, and Forge-based UIs to rocm7.2 / cu128 depending on GPU - thanks to @NeuralFault!
25
-
### Fixed
26
-
- Fixed the Package Manager "Add Package" teaching tip opening inopportunely while packages were still loading or after opening the add-package dialog
27
-
- Fixed downloaded checkpoint going to StableDiffusion folder when a saved download preference existed, even for GGUF files that should always go to DiffusionModels
28
-
- Fixed potential crash when adding metadata to malformed or non-PNG image data in Inference
29
-
- Fixed non-Latin-1 characters (e.g. Japanese, Chinese, Korean, emoji) in image generation parameters being stored in PNG tEXt chunks, violating the PNG specification and causing character corruption (mojibake) in standard-compliant parsers. Non-Latin-1 content now uses spec-compliant iTXt chunks with proper UTF-8 encoding ([#1535](https://github.com/LykosAI/StabilityMatrix/issues/1535))
30
-
- Fixed an issue where `Align Your Steps` scheduler and Unet Loader workflows ignored Regional Prompting (and other addon) conditioning modifiers.
31
-
- Fixed bold text not rendering in markdown dialogs on Windows 11 due to Avalonia 11.3.x variable font regression with Segoe UI Variable Text
32
-
- Fixed Japanese text appearing compressed/squished in markdown dialogs by ensuring the bundled NotoSansJP font is used for CTextBlock rendering
33
-
- Fixed ContentDialog title and buttons not using the correct font for Japanese locale (NotoSansJP) when shown as overlay
34
-
- Added missing `CBold` and `CItalic` inline styles to the markdown style sheet
35
-
- Fixed downloads failing with "The request message was already sent" when the server doesn't return Content-Length on the first attempt, caused by reusing a consumed HttpRequestMessage in the retry loop
36
-
- Fixed downloads from sources that redirect to CivitAI/HuggingFace (e.g. CivArchive) failing with Unauthorized by resolving the redirect target URL and applying auth headers for the correct domain
37
-
- Fixed dropdown menu overlayed in Inference UI Model Cards not being scrollable on Linux - thanks to @NeuralFault!
38
-
- Fixed model downloads failing on VPN connections - thanks to @NeuralFault!
39
-
- Fixed [#1598](https://github.com/LykosAI/StabilityMatrix/issues/1598) - download progress bar showing 100% immediately for fresh downloads due to missing Content-Length fallback when Content-Range header is absent
40
-
- Fixed [#1597](https://github.com/LykosAI/StabilityMatrix/issues/1597) - reForge launch failing due to setuptools version
41
-
- Fixed [#1596](https://github.com/LykosAI/StabilityMatrix/issues/1596) - package installs and managed embedded Python startup being poisoned by inherited shell Python activation variables such as `PYTHONHOME`, `PYTHONPATH`, `VIRTUAL_ENV`, and Conda environment variables
42
-
- Fixed [#1590](https://github.com/LykosAI/StabilityMatrix/issues/1590) - Startup crash when settings file is corrupted. Settings files are now self-healing with automatic recovery from null bytes, truncated JSON, and missing brackets
43
-
- Potentially fixed [#1578](https://github.com/LykosAI/StabilityMatrix/issues/1578) - `SocketException: Address already in use` on Linux startup by cleaning stale interprocess socket files and reactivating the existing window
44
-
- Fixed [#1397](https://github.com/LykosAI/StabilityMatrix/issues/1397), [#610](https://github.com/LykosAI/StabilityMatrix/issues/610) - duplicate pip package entries in results - thanks to @e-nord!
45
-
### Supporters
46
-
#### 🌟 Visionaries
47
-
A heartfelt thank you to our incredible Visionaries: **Waterclouds**, **JungleDragon**, **bluepopsicle**, **Bob S**, and **whudunit** - every feature, fix, and late-night breakthrough in this release carries your fingerprints. A huge welcome to our newest Visionaries **Droolguy** and **snotty** (leveling up from the Pioneer ranks!), a warm welcome back to longtime Visionary **Ibixat**, and an equally huge welcome to **LG**, making their Stability Matrix debut straight at the Visionary tier! You're the reason we can keep building bold things - and an extra-special thank you to everyone now supporting us directly through our new platform. Your trust in this next chapter means the world!
48
-
49
-
## v2.16.0-dev.2
50
-
### Added
51
-
- Added Regional Prompting addon to Inference - paint detailed masks to apply different prompts, strengths, and settings to specific regions of your image
52
-
- Multi-layer mask editor with Photoshop-style interface for managing layers with independent masks, prompts, colors, and opacity
53
-
- Professional brush tools: freehand brush/eraser with pressure sensitivity, rectangle/ellipse shapes with fill/stroke modes, paint bucket flood fill
54
-
- Brush feathering/softness control for smooth, blended mask edges (0 = hard edge, 1 = soft/blurred)
55
-
- Per-layer prompt and strength controls, export/import masks as PNG, duplicate layers, image reference layers for tracing
56
-
- GPU-accelerated rendering with compact gzip-compressed metadata serialization
57
-
- Added new Model Picker dialog for Inference with grid/list views, search, filtering, and NSFW overlay
58
-
- Added browse buttons to all model dropdowns in Inference (Model, Refiner, VAE, Text Encoders, CLIP Vision)
59
-
- Added inline search box to model combo box dropdowns with fuzzy matching
60
-
- Added NVIDIA driver version warning when launching ComfyUI with CUDA 13.0 (cu130) and driver versions below 580.x
61
-
- Added legacy Python warning when launching InvokeAI installations using Python 3.10.11
62
-
- Added Tiled VAE Decode to the Inference video workflows - thanks to @NeuralFault!
63
-
- Added recoverable error dialog for UI thread exceptions, with option to continue instead of exiting
64
-
### Changed
65
-
- Disabled update checking for legacy InvokeAI installations using Python 3.10.11
66
-
- Hide rating stars in the Civitai browser page if no rating is available
67
-
- Updated uv to v0.9.30
68
-
- Updated PortableGit to v2.52.0.windows.1
69
-
- Updated Sage/Triton/Nunchaku installers to use GitHub API to fetch latest releases
70
-
- Updated ComfyUI installations and updates to automatically install ComfyUI Manager
71
-
- Updated gfx110X Windows ROCm nightly index - thanks to @NeuralFault!
72
-
- Updated ComfyUI-Zluda install to more closely match the author's intended installation method - thanks to @NeuralFault!
73
-
- Updated Forge Classic installs/updates to use the upstream install script for better version compatibility with torch/sage/triton/nunchaku
74
-
### Fixed
75
-
- Fixed parsing of escape sequences in Inference such as `\\`
76
-
- Fixed batch notification firing when only one image is generated
77
-
- Fixed [#1546](https://github.com/LykosAI/StabilityMatrix/issues/1546), [#1541](https://github.com/LykosAI/StabilityMatrix/issues/1541) - "No module named 'pkg_resources'" error when installing Automatic1111/Forge/reForge packages
- Fixed [#1529](https://github.com/LykosAI/StabilityMatrix/issues/1529) - "Selected commit is null" error when installing packages and rate limited by GitHub
80
-
- Fixed [#1525](https://github.com/LykosAI/StabilityMatrix/issues/1525) - Crash after downloading a model
81
-
- Fixed [#1523](https://github.com/LykosAI/StabilityMatrix/issues/1523), [#1499](https://github.com/LykosAI/StabilityMatrix/issues/1499), [#1494](https://github.com/LykosAI/StabilityMatrix/issues/1494) - Automatic1111 using old stable diffusion repo
82
-
- Fixed [#1505](https://github.com/LykosAI/StabilityMatrix/issues/1505) - incorrect port argument for Wan2GP
83
-
- Possibly fix [#1502](https://github.com/LykosAI/StabilityMatrix/issues/1502) - English fonts not displaying correctly on Linux in Chinese environments
- Fixed [#1466](https://github.com/LykosAI/StabilityMatrix/issues/1466) - crash after moving portable install
86
-
- Fixed [#1445](https://github.com/LykosAI/StabilityMatrix/issues/1445) - Linux app updates not actually updating - thanks to @NeuralFault!
87
-
### Supporters
88
-
#### 🌟 Visionaries
89
-
Huge shoutout to our amazing Visionaries: **Waterclouds**, **JungleDragon**, **bluepopsicle**, **Bob S**, and **whudunit**! Your continued support fuels every new feature and improvement in Stability Matrix. We couldn't do it without you - thank you for believing in what we're building!
90
-
91
-
## v2.16.0-dev.1
92
-
### Added
93
-
#### New Feature: 🧪 Image Lab - Conversational Image Generation for ComfyUI
94
-
- We've added a brand new conversational interface for image generation! Image Lab lets you iterate on images naturally through chat, rather than just one-off prompts.
95
-
- Local-First Power: Native support for Flux Kontext and Qwen Image Edit running entirely locally via your ComfyUI backend.
96
-
- Smart Setup: Stability Matrix automatically detects and helps you download the specific models and LoRAs needed for these local workflows.
97
-
- Interactive Tools: Drag-and-drop image inputs, use the built-in annotation tool to draw on images, and keep persistent conversation history.
98
-
- Cloud Option: Includes optional support for Nano Banana (Gemini 3 Pro/2.5) for users who want to leverage external reasoning models.
99
-
- Added new package - [Wan2GP](https://github.com/deepbeepmeep/Wan2GP)
100
-
- Added [Stable Diffusion WebUI Forge - Neo](https://github.com/Haoming02/sd-webui-forge-classic/tree/neo) as a separate package for convenience
101
-
- Added Intel GPU support for ComfyUI
102
-
- Added "Run Python Command" option to the package card's 3-dots menu for running arbitrary Python code in the package's virtual environment
103
-
- Added togglable `--uv` argument to the SD.Next launch options
104
-
- Added Tiled VAE decoding as an Inference addon thanks to @NeuralFault!
105
-
### Changed
106
-
- Moved the original Stable Diffusion WebUI Forge to the "Legacy" packages tab due to inactivity
107
-
- Updated to cu130 torch index for ComfyUI installs with Nvidia GPUs
108
-
- Consolidated and fixed AMD GPU architecture detection
109
-
- Updated SageAttention installer to latest v2.2.0-windows.post4 version
110
-
- Video files can now be opened directly from the Output browser
111
-
- Videos will now appear with thumbnails in the Output browser
112
-
### Fixed
113
-
- Fixed [#1450](https://github.com/LykosAI/StabilityMatrix/issues/1450) - Older SD.Next not launching due to forced `--uv` argument
114
-
- Fixed duplicate custom node installations when installing workflows from the Workflow Browser - thanks again to @NeuralFault!
115
-
### Supporters
116
-
#### 🌟 Visionaries
117
-
A massive thank you to our esteemed Visionaries: **Waterclouds**, **JungleDragon**, **bluepopsicle**, **Bob S**, and **whudunit**! Your generosity is the powerhouse behind Stability Matrix, enabling us to keep building and refining with confidence. We are truly grateful for your partnership!
118
-
119
-
>>>>>>> ce8c057d (Merge pull request #1230 from ionite34/shoutout-chagenlog)
120
8
## v2.15.7
121
9
### Added
122
10
- Added single-instance window activation signaling so reopening the app restores and focuses the existing desktop window instead of launching a duplicate instance
0 commit comments