Text-to-Image
Diffusers
lora
patrickvonplaten commited on
Commit
0f5db1b
1 Parent(s): 8e48c6c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -8,12 +8,12 @@ license: openrail++
8
  inference: false
9
  ---
10
 
11
- # Latent Consistency Model (LCM) LoRA: SDv1-5
12
 
13
  Latent Consistency Model (LCM) LoRA was proposed in [LCM-LoRA: A universal Stable-Diffusion Acceleration Module](TODO:)
14
  by *Simian Luo, Yiqin Tan, Suraj Patil, Daniel Gu et al.*
15
 
16
- It is a distilled consistency adapter for [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) that allows
17
  to reduce the number of inference steps to only between **2 - 8 steps**.
18
 
19
  | Model | Params / M |
@@ -35,14 +35,14 @@ pip install --upgrade diffusers transformers accelerate peft
35
 
36
  ### Text-to-Image
37
 
38
- The adapter can be loaded with SDv1-5 or deviratives. Here we use [`Lykon/dreamshaper-7`](https://huggingface.co/Lykon/dreamshaper-7). Next, the scheduler needs to be changed to [`LCMScheduler`](https://huggingface.co/docs/diffusers/v0.22.3/en/api/schedulers/lcm#diffusers.LCMScheduler) and we can reduce the number of inference steps to just 2 to 8 steps.
39
  Please make sure to either disable `guidance_scale` or use values between 1.0 and 2.0.
40
 
41
  ```python
42
  import torch
43
  from diffusers import LCMScheduler, AutoPipelineForText2Image
44
 
45
- model_id = "Lykon/dreamshaper-7"
46
  adapter_id = "latent-consistency/lcm-lora-sdv1-5"
47
 
48
  pipe = AutoPipelineForText2Image.from_pretrained(model_id, torch_dtype=torch.float16, variant="fp16")
 
8
  inference: false
9
  ---
10
 
11
+ # Latent Consistency Model (LCM) LoRA: SSD-1B
12
 
13
  Latent Consistency Model (LCM) LoRA was proposed in [LCM-LoRA: A universal Stable-Diffusion Acceleration Module](TODO:)
14
  by *Simian Luo, Yiqin Tan, Suraj Patil, Daniel Gu et al.*
15
 
16
+ It is a distilled consistency adapter for [`segmind/SSD-1B`](https://huggingface.co/runwayml/stable-diffusion-v1-5) that allows
17
  to reduce the number of inference steps to only between **2 - 8 steps**.
18
 
19
  | Model | Params / M |
 
35
 
36
  ### Text-to-Image
37
 
38
+ Let's load the base model `segmind/SSD-1B` first. Next, the scheduler needs to be changed to [`LCMScheduler`](https://huggingface.co/docs/diffusers/v0.22.3/en/api/schedulers/lcm#diffusers.LCMScheduler) and we can reduce the number of inference steps to just 2 to 8 steps.
39
  Please make sure to either disable `guidance_scale` or use values between 1.0 and 2.0.
40
 
41
  ```python
42
  import torch
43
  from diffusers import LCMScheduler, AutoPipelineForText2Image
44
 
45
+ model_id = "segmind/SSD-1B"
46
  adapter_id = "latent-consistency/lcm-lora-sdv1-5"
47
 
48
  pipe = AutoPipelineForText2Image.from_pretrained(model_id, torch_dtype=torch.float16, variant="fp16")