Update README.md
Browse files
README.md
CHANGED
|
@@ -16,7 +16,6 @@ waifu is a free text-to-image model that can efficiently generate images in 80 l
|
|
| 16 |
(2) [**Linear DiT**](https://github.com/NVlabs/Sana): we use 1.6b DiT transformer with linear attention. \
|
| 17 |
(3) [**MEXMA-SigLIP**](https://huggingface.co/visheratin/mexma-siglip): MEXMA-SigLIP is a model that combines the [MEXMA](https://huggingface.co/facebook/MEXMA) multilingual text encoder and an image encoder from the [SigLIP](https://huggingface.co/timm/ViT-SO400M-14-SigLIP-384) model. This allows us to get a high-performance CLIP model for 80 languages.. \
|
| 18 |
(4) Other: we use Flow-Euler sampler, Adafactor-Fused optimizer and bf16 precision for training, and combine efficient caption labeling (MoonDream, CogVlM, Human, Gpt's) and danbooru tags to accelerate convergence.
|
| 19 |
-
,
|
| 20 |
|
| 21 |
## Pros
|
| 22 |
- Small model that can be trained on a common GPU; fast training process.
|
|
|
|
| 16 |
(2) [**Linear DiT**](https://github.com/NVlabs/Sana): we use 1.6b DiT transformer with linear attention. \
|
| 17 |
(3) [**MEXMA-SigLIP**](https://huggingface.co/visheratin/mexma-siglip): MEXMA-SigLIP is a model that combines the [MEXMA](https://huggingface.co/facebook/MEXMA) multilingual text encoder and an image encoder from the [SigLIP](https://huggingface.co/timm/ViT-SO400M-14-SigLIP-384) model. This allows us to get a high-performance CLIP model for 80 languages.. \
|
| 18 |
(4) Other: we use Flow-Euler sampler, Adafactor-Fused optimizer and bf16 precision for training, and combine efficient caption labeling (MoonDream, CogVlM, Human, Gpt's) and danbooru tags to accelerate convergence.
|
|
|
|
| 19 |
|
| 20 |
## Pros
|
| 21 |
- Small model that can be trained on a common GPU; fast training process.
|