Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-19 17:24:32 +02:00
Code Issues Packages Projects Releases Wiki Activity

update 

Signed-off-by: Vladimir Mandic <mandic00@live.com>
Vladimir Mandic
2025-05-15 09:00:49 -04:00
parent e3b0583b65
commit 6192bb85f1
4 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
AMD-ROCm.md
+1 -1
@@ -123,7 +123,7 @@ On the other hand, for best performance during generation (but slower startup on
### Reduce VRAM consumption
If you use the `bf16` data type (*Settings > Compute Settings > Execution Precision > Device precision type*), which is autodetected on RDNA3 and newer cards, there is the chance that VRAM usage will be very high (16+ GB) when decoding the final image and when upscaling with non-latent upscalers. To workaround the problem, ensure to set `Device precision type` as `fp16`, and disable VAE upcasting in *Variable Auto Encoder > VAE upcasting*.
If you use the `bf16` data type (*Settings > Compute Settings > Execution Precision > Device precision type*), which is autodetected on RDNA3 and newer cards, there is the chance that VRAM usage will be very high (16+ GB) when decoding the final image and when upscaling with non-latent upscalers. To workaround the problem, ensure to set `Device precision type` as `fp16`, and disable VAE upcasting in *Variational Auto Encoder > VAE upcasting*.
Setting `fp16` has also a noticeable impact on performance.
### Composable Kernel (CK) Flash attention
CHANGELOG.md
+1 -1
@@ -875,7 +875,7 @@ Commit hash: `master: #dcfc9f3` `dev: #935cac6`
- optimizations: full offload, quantization and tiling support
- [TeaCache](https://github.com/ali-vilab/TeaCache/blob/main/TeaCache4LTX-Video/README.md) integration
- **VAE**:
- tiling granular options in *settings -> variable auto encoder*
- tiling granular options in *settings -> Variational Auto Encoder*
- **UI**:
- live preview optimizations and error handling
- live preview high quality output, thanks @Disty0
Home.md
+1 -1
@@ -74,7 +74,7 @@ Learn how you can manage your resources better:
- [Memory Offloading](Offload)
includes performance notes for different offload options
- [Model Quantization](Quantization)
- [Variable Auto-Encoder](VAE)
- [Variational Autoencoder](VAE)
- [Nunchaku](Nunchaku) how to use for faster inference
Plus collection of [Benchmarks](Benchmark) to highlight what to expect using different compile and/or device settings: - Compile/device settings
VAE.md
+1 -1
@@ -1,4 +1,4 @@
# Variable Auto-Encoder (VAE)
# Variational Autoencoder (VAE)
VAE is a model that can be used to compress and decompress images.
It is a type of autoencoder that learns a latent space representation of the input data. The model is trained to minimize the reconstruction error of the input data, while also learning a latent space that is continuous and smooth. This allows the model to generate new images by sampling from the latent space.