Instructions to use PrunaAI/Pruna-Qwen-Image-2.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use PrunaAI/Pruna-Qwen-Image-2.1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Qwen/Qwen-Image-2.1", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("PrunaAI/Pruna-Qwen-Image-2.1") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Fix cloning issue
Browse files
README.md
CHANGED
|
@@ -87,7 +87,7 @@ STEPS = 8 # 8 for higher quality, 5 for higher speed
|
|
| 87 |
|
| 88 |
# The terminal sigma 0 is appended by the scheduler.
|
| 89 |
SIGMAS = {
|
| 90 |
-
5: [1.0,
|
| 91 |
8: [1.0, 14 / 15, 6 / 7, 10 / 13, 2 / 3, 6 / 11, 0.4, 2 / 9],
|
| 92 |
}[STEPS]
|
| 93 |
|
|
@@ -144,7 +144,7 @@ image.save("edit_example.png")
|
|
| 144 |
at `shift=1.0` with dynamic shifting off so the sigmas are not shifted twice.
|
| 145 |
- 8-step: `1 β 14/15 β 6/7 β 10/13 β 2/3 β 6/11 β 0.4 β 2/9 β 0`
|
| 146 |
(shift 2, computed as `Ο = 2t / (1 + t)` on evenly spaced `t`)
|
| 147 |
-
- 5-step: `1 β
|
| 148 |
- **No CFG.** Keep `true_cfg_scale=1.0` and do not pass a negative prompt.
|
| 149 |
- **Keep the LoRA strength at 1.0.**
|
| 150 |
- **Write detailed prompts for text-to-image.** Longer prompts that describe
|
|
|
|
| 87 |
|
| 88 |
# The terminal sigma 0 is appended by the scheduler.
|
| 89 |
SIGMAS = {
|
| 90 |
+
5: [1.0, 0.94, 6 / 7, 2 / 3, 0.4],
|
| 91 |
8: [1.0, 14 / 15, 6 / 7, 10 / 13, 2 / 3, 6 / 11, 0.4, 2 / 9],
|
| 92 |
}[STEPS]
|
| 93 |
|
|
|
|
| 144 |
at `shift=1.0` with dynamic shifting off so the sigmas are not shifted twice.
|
| 145 |
- 8-step: `1 β 14/15 β 6/7 β 10/13 β 2/3 β 6/11 β 0.4 β 2/9 β 0`
|
| 146 |
(shift 2, computed as `Ο = 2t / (1 + t)` on evenly spaced `t`)
|
| 147 |
+
- 5-step: `1 β 0.94 β 6/7 β 2/3 β 0.4 β 0`
|
| 148 |
- **No CFG.** Keep `true_cfg_scale=1.0` and do not pass a negative prompt.
|
| 149 |
- **Keep the LoRA strength at 1.0.**
|
| 150 |
- **Write detailed prompts for text-to-image.** Longer prompts that describe
|