Skip to content

Commit 0a19f9b

Browse files
JeffreyXiangJeffreyXiang
committed
Update readme
1 parent 7b96b73 commit 0a19f9b

File tree

1 file changed

+15
-43
lines changed

1 file changed

+15
-43
lines changed

‎README.md

Lines changed: 15 additions & 43 deletions
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22
<h1 align="center">Structured 3D Latents<br>for Scalable and Versatile 3D Generation</h1>
33
<p align="center"><a href="https://arxiv.org/abs/2412.01506"><img src='https://img.shields.io/badge/arXiv-Paper-red?logo=arxiv&logoColor=white' alt='arXiv'></a>
44
<a href='https://trellis3d.github.io'><img src='https://img.shields.io/badge/Project_Page-Website-green?logo=googlechrome&logoColor=white' alt='Project Page'></a>
5-
<a href='https://huggingface.co/spaces/JeffreyXiang/TRELLIS'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Live_Demo-blue'></a>
5+
<a href='https://huggingface.co/spaces?q=TRELLIS'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Live_Demo-blue'></a>
66
</p>
77
<p align="center"><img src="assets/teaser.png" width="100%"></p>
88

@@ -84,32 +84,6 @@
8484
--demo Install all dependencies for demo
8585
```
8686

87-
<!-- Pretrained Models -->
88-
## 🤖 Pretrained Models
89-
90-
We provide the following pretrained models:
91-
92-
| Model | Description | #Params | Download |
93-
| --- | --- | --- | --- |
94-
| TRELLIS-image-large | Large image-to-3D model | 1.2B | [Download](https://huggingface.co/microsoft/TRELLIS-image-large) |
95-
| TRELLIS-text-base | Base text-to-3D model | 342M | [Download](https://huggingface.co/microsoft/TRELLIS-text-base) |
96-
| TRELLIS-text-large | Large text-to-3D model | 1.1B | [Download](https://huggingface.co/microsoft/TRELLIS-text-large) |
97-
| TRELLIS-text-xlarge | Extra-large text-to-3D model | 2.0B | [Download](https://huggingface.co/microsoft/TRELLIS-text-xlarge) |
98-
99-
*Note: It is always recommended to use the image conditioned version of the models for better performance.*
100-
101-
*Note: All VAEs are included in **TRELLIS-image-large** model repo.*
102-
103-
The models are hosted on Hugging Face. You can directly load the models with their repository names in the code:
104-
```python
105-
TrellisImageTo3DPipeline.from_pretrained("microsoft/TRELLIS-image-large")
106-
```
107-
108-
If you prefer loading the model from local, you can download the model files from the links above and load the model with the folder path (folder structure should be maintained):
109-
```python
110-
TrellisImageTo3DPipeline.from_pretrained("/path/to/TRELLIS-image-large")
111-
```
112-
11387
<!-- Usage -->
11488
## 💡 Usage
11589

@@ -199,8 +173,6 @@ python app.py
199173

200174
Then, you can access the demo at the address shown in the terminal.
201175

202-
***The web demo is also available on [Hugging Face Spaces](https://huggingface.co/spaces/JeffreyXiang/TRELLIS)!***
203-
204176

205177
<!-- Dataset -->
206178
## 📚 Dataset
@@ -239,20 +211,20 @@ TRELLIS’s training framework is organized to provide a flexible and modular ap
239211
- Training hyperparameters and model architectures are defined in configuration files under the `configs/` directory.
240212
- Example configuration files include:
241213

242-
| Config | Pretained Model | Description |
243-
| --- | --- | --- |
244-
| [`vae/ss_vae_conv3d_16l8_fp16.json`](configs/vae/ss_vae_conv3d_16l8_fp16.json) | [Encoder](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/ss_enc_conv3d_16l8_fp16.safetensors) [Decoder](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/ss_dec_conv3d_16l8_fp16.safetensors) | Sparse structure VAE |
245-
| [`vae/slat_vae_enc_dec_gs_swin8_B_64l8_fp16.json`](configs/vae/slat_vae_enc_dec_gs_swin8_B_64l8_fp16.json) | [Encoder](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/slat_enc_swin8_B_64l8_fp16.safetensors) [Decoder](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/slat_dec_gs_swin8_B_64l8gs32_fp16.safetensors) | SLat VAE with Gaussian Decoder |
246-
| [`vae/slat_vae_dec_rf_swin8_B_64l8_fp16.json`](configs/vae/slat_vae_dec_rf_swin8_B_64l8_fp16.json) | [Decoder](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/slat_dec_rf_swin8_B_64l8r16_fp16.safetensors) | SLat Radiance Field Decoder |
247-
| [`vae/slat_vae_dec_mesh_swin8_B_64l8_fp16.json`](configs/vae/slat_vae_dec_mesh_swin8_B_64l8_fp16.json) | [Decoder](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/slat_dec_mesh_swin8_B_64l8m256c_fp16.safetensors) | SLat Mesh Decoder |
248-
| [`generation/ss_flow_img_dit_L_16l8_fp16.json`](configs/generation/ss_flow_img_dit_L_16l8_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/ss_flow_img_dit_L_16l8_fp16.safetensors) | Image conditioned sparse structure Flow Model |
249-
| [`generation/slat_flow_img_dit_L_64l8p2_fp16.json`](configs/generation/slat_flow_img_dit_L_64l8p2_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-image-large/blob/main/ckpts/slat_flow_img_dit_L_64l8p2_fp16.safetensors) | Image conditioned SLat Flow Model |
250-
| [`generation/ss_flow_txt_dit_B_16l8_fp16.json`](configs/generation/ss_flow_txt_dit_B_16l8_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-text-base/blob/main/ckpts/ss_flow_txt_dit_B_16l8_fp16.safetensors) | Base text-conditioned sparse structure Flow Model |
251-
| [`generation/slat_flow_txt_dit_B_64l8p2_fp16.json`](configs/generation/slat_flow_txt_dit_B_64l8p2_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-text-base/blob/main/ckpts/slat_flow_txt_dit_B_64l8p2_fp16.safetensors) | Base text-conditioned SLat Flow Model |
252-
| [`generation/ss_flow_txt_dit_L_16l8_fp16.json`](configs/generation/ss_flow_txt_dit_L_16l8_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-text-large/blob/main/ckpts/ss_flow_txt_dit_L_16l8_fp16.safetensors) | Large text-conditioned sparse structure Flow Model |
253-
| [`generation/slat_flow_txt_dit_L_64l8p2_fp16.json`](configs/generation/slat_flow_txt_dit_L_64l8p2_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-text-large/blob/main/ckpts/slat_flow_txt_dit_L_64l8p2_fp16.safetensors) | Large text-conditioned SLat Flow Model |
254-
| [`generation/ss_flow_txt_dit_XL_16l8_fp16.json`](configs/generation/ss_flow_txt_dit_XL_16l8_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-text-xlarge/blob/main/ckpts/ss_flow_txt_dit_XL_16l8_fp16.safetensors) | Extra-large text-conditioned sparse structure Flow Model |
255-
| [`generation/slat_flow_txt_dit_XL_64l8p2_fp16.json`](configs/generation/slat_flow_txt_dit_XL_64l8p2_fp16.json) | [Denoiser](https://huggingface.co/microsoft/TRELLIS-text-xlarge/blob/main/ckpts/slat_flow_txt_dit_XL_64l8p2_fp16.safetensors) | Extra-large text-conditioned SLat Flow Model |
214+
| Config | Description |
215+
| --- | --- |
216+
| [`vae/ss_vae_conv3d_16l8_fp16.json`](configs/vae/ss_vae_conv3d_16l8_fp16.json) | Sparse structure VAE |
217+
| [`vae/slat_vae_enc_dec_gs_swin8_B_64l8_fp16.json`](configs/vae/slat_vae_enc_dec_gs_swin8_B_64l8_fp16.json) | SLat VAE with Gaussian Decoder |
218+
| [`vae/slat_vae_dec_rf_swin8_B_64l8_fp16.json`](configs/vae/slat_vae_dec_rf_swin8_B_64l8_fp16.json) | SLat Radiance Field Decoder |
219+
| [`vae/slat_vae_dec_mesh_swin8_B_64l8_fp16.json`](configs/vae/slat_vae_dec_mesh_swin8_B_64l8_fp16.json) | SLat Mesh Decoder |
220+
| [`generation/ss_flow_img_dit_L_16l8_fp16.json`](configs/generation/ss_flow_img_dit_L_16l8_fp16.json) | Image conditioned sparse structure Flow Model |
221+
| [`generation/slat_flow_img_dit_L_64l8p2_fp16.json`](configs/generation/slat_flow_img_dit_L_64l8p2_fp16.json) | Image conditioned SLat Flow Model |
222+
| [`generation/ss_flow_txt_dit_B_16l8_fp16.json`](configs/generation/ss_flow_txt_dit_B_16l8_fp16.json) | Base text-conditioned sparse structure Flow Model |
223+
| [`generation/slat_flow_txt_dit_B_64l8p2_fp16.json`](configs/generation/slat_flow_txt_dit_B_64l8p2_fp16.json) | Base text-conditioned SLat Flow Model |
224+
| [`generation/ss_flow_txt_dit_L_16l8_fp16.json`](configs/generation/ss_flow_txt_dit_L_16l8_fp16.json) | Large text-conditioned sparse structure Flow Model |
225+
| [`generation/slat_flow_txt_dit_L_64l8p2_fp16.json`](configs/generation/slat_flow_txt_dit_L_64l8p2_fp16.json) | Large text-conditioned SLat Flow Model |
226+
| [`generation/ss_flow_txt_dit_XL_16l8_fp16.json`](configs/generation/ss_flow_txt_dit_XL_16l8_fp16.json) | Extra-large text-conditioned sparse structure Flow Model |
227+
| [`generation/slat_flow_txt_dit_XL_64l8p2_fp16.json`](configs/generation/slat_flow_txt_dit_XL_64l8p2_fp16.json) | Extra-large text-conditioned SLat Flow Model |
256228

257229

258230
### Command-Line Options

0 commit comments

Comments
 (0)