diff --git a/changelog/index.mdx b/changelog/index.mdx
index ca3a706a5..dce46b499 100644
--- a/changelog/index.mdx
+++ b/changelog/index.mdx
@@ -4,6 +4,37 @@ description: "Track ComfyUI's latest features, improvements, and bug fixes"
 icon: "clock-rotate-left"
 ---
 
+<Update label="v0.3.33" description="July 9, 2025">
+
+**ACE Audio Model Support & Enhanced Memory Management**
+
+This release introduces comprehensive support for ACE (Audio Conditioning Enhancement) models and delivers important memory optimizations for large-scale workflows:
+
+## New ACE Model Integration
+- **ACE-Step Model Support**: Initial implementation brings audio-conditioned generation capabilities to ComfyUI, enabling music and sound-aware image generation workflows
+- **Experimental Lyrics Strength**: Fine-tune how lyrical content influences generation through new strength controls, perfect for music-themed creative workflows
+- **Japanese Character Support**: Full hiragana and katakana character compatibility ensures ACE models work seamlessly with Japanese text inputs
+- **ACE VAE Tiling**: Efficient tiling support for ACE VAE processing, enabling high-resolution audio-conditioned generation without memory constraints
+
+## Memory & Performance Optimizations
+- **Enhanced LTXV 13B Estimation**: Improved memory estimation for the massive 13B parameter LTXV model, preventing out-of-memory errors in complex workflows
+- **ACE VAE Memory Tuning**: Optimized memory allocation for ACE VAE operations, supporting larger batch sizes and higher resolution processing
+- **LTXV Config Detection**: Automatic detection code ensures LTXV models work even without explicit configuration files
+
+## Workflow Reliability Improvements
+- **Flux Shift Integration**: Chroma now uses Flux shift for more consistent color handling across different model types
+- **PyAV Version Protection**: Prevents loading of API nodes on incompatible PyAV versions, avoiding runtime errors in video processing workflows
+- **Optional TorchAudio**: Removed hard torchaudio requirement, reducing installation complexity for users focused on image generation
+
+## Developer Experience Enhancements
+- **Flexible Dependencies**: More permissive dependency management reduces installation friction for custom node developers
+- **Better Error Handling**: Improved error messages and validation for audio processing components
+- **Compatibility Layers**: Enhanced backward compatibility for existing workflows using audio features
+
+These improvements make ComfyUI more accessible for audio-visual creative workflows while maintaining stability for traditional image generation tasks. The ACE model support opens new possibilities for music-driven AI art creation and audio-conditioned generation workflows.
+
+</Update>
+
 <Update label="v0.3.44" description="July 8, 2025">
 
 **Advanced Sampling & Model Control Enhancements**
@@ -93,7 +124,6 @@ These changes enhance the reliability of advanced model workflows while maintain
 
 This release significantly expands ComfyUI's model ecosystem support while delivering crucial stability improvements and enhanced hardware compatibility across different platforms.
 </Update>
-
 <Update label="v0.3.40" description="June 5, 2025">
 
 This release brings powerful new workflow utilities and performance optimizations for ComfyUI creators:
diff --git a/zh-CN/changelog/index.mdx b/zh-CN/changelog/index.mdx
index 4715759d3..e4fcfcf99 100644
--- a/zh-CN/changelog/index.mdx
+++ b/zh-CN/changelog/index.mdx
@@ -4,6 +4,36 @@ description: "跟踪ComfyUI的最新功能、改进和错误修复"
 icon: "clock-rotate-left"
 ---
 
+<Update label="v0.3.33" description="2025年7月9日">
+
+**ACE音频模型支持和增强的内存管理**
+
+此版本引入了对ACE（音频条件增强）模型的全面支持，并为大规模工作流程提供了重要的内存优化：
+
+## 新的ACE模型集成
+- **ACE-Step模型支持**：初始实现为ComfyUI带来音频条件生成功能，支持音乐和声音感知的图像生成工作流程
+- **实验性歌词强度**：通过新的强度控制精细调整歌词内容对生成的影响，非常适合音乐主题的创意工作流程
+- **日文字符支持**：全面兼容平假名和片假名字符，确保ACE模型与日文文本输入无缝协作
+- **ACE VAE分块**：为ACE VAE处理提供高效的分块支持，在无内存限制的情况下实现高分辨率音频条件生成
+
+## 内存和性能优化
+- **增强的LTXV 13B估算**：改进了对大规模130亿参数LTXV模型的内存估算，防止复杂工作流程中出现内存不足错误
+- **ACE VAE内存调优**：优化了ACE VAE操作的内存分配，支持更大的批处理大小和更高分辨率的处理
+- **LTXV配置检测**：自动检测代码确保即使没有显式配置文件，LTXV模型也能正常工作
+
+## 工作流程可靠性改进
+- **Flux Shift集成**：Chroma现在使用Flux shift在不同模型类型之间实现更一致的颜色处理
+- **PyAV版本保护**：防止在不兼容的PyAV版本上加载API节点，避免视频处理工作流程中的运行时错误
+- **可选TorchAudio**：移除硬性的torchaudio要求，为专注于图像生成的用户降低安装复杂性
+
+## 开发者体验增强
+- **灵活的依赖关系**：更宽松的依赖管理减少了自定义节点开发者的安装阻力
+- **更好的错误处理**：改进了音频处理组件的错误消息和验证
+- **兼容性层**：增强了使用音频功能的现有工作流程的向后兼容性
+
+这些改进使ComfyUI在音频视觉创意工作流程中更加易用，同时保持了传统图像生成任务的稳定性。ACE模型支持为音乐驱动的AI艺术创作和音频条件生成工作流程开辟了新的可能性。
+
+</Update>
 <Update label="v0.3.44" description="2025年7月8日">
 
 **高级采样和模型控制增强**