Instructions to use dataautogpt3/PrometheusV1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use dataautogpt3/PrometheusV1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("dataautogpt3/PrometheusV1", dtype=torch.bfloat16, device_map="cuda") prompt = "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Update README.md
Browse files
README.md
CHANGED
|
@@ -64,11 +64,9 @@ Prometheus is presumed to be the first full rank finetune of Playground v2.5, de
|
|
| 64 |
|
| 65 |
# Key Features and Considerations
|
| 66 |
|
| 67 |
-
Presumed First Full Rank Finetune of Playground v2.5:
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
Unique approach to fine-tuning this particular base model
|
| 71 |
-
|
| 72 |
|
| 73 |
# Enhanced Accessibility:
|
| 74 |
|
|
@@ -78,21 +76,19 @@ Designed to be more compatible with standard open-source tools and workflows
|
|
| 78 |
|
| 79 |
# Output Characteristics:
|
| 80 |
|
| 81 |
-
Aims to provide a balance between consistency and diversity in outputs
|
| 82 |
-
May exhibit some stylistic tendencies inherited from the training process
|
| 83 |
-
|
| 84 |
|
| 85 |
# Training Approach:
|
| 86 |
|
| 87 |
-
Utilizes the extensive Proteus datasets
|
| 88 |
-
Focused on maintaining model capabilities while increasing compatibility
|
| 89 |
|
| 90 |
|
| 91 |
# Custom CLIP Integration:
|
| 92 |
|
| 93 |
-
Incorporation of a specially trained CLIP model
|
| 94 |
-
Requires a clip skip setting of 2 for optimal performance
|
| 95 |
-
|
| 96 |
|
| 97 |
|
| 98 |
# About Prometheus
|
|
|
|
| 64 |
|
| 65 |
# Key Features and Considerations
|
| 66 |
|
| 67 |
+
- Presumed First Full Rank Finetune of Playground v2.5:
|
| 68 |
+
- Complete parameter update of Playground v2.5 architecture
|
| 69 |
+
- Unique approach to fine-tuning this particular base model
|
|
|
|
|
|
|
| 70 |
|
| 71 |
# Enhanced Accessibility:
|
| 72 |
|
|
|
|
| 76 |
|
| 77 |
# Output Characteristics:
|
| 78 |
|
| 79 |
+
- Aims to provide a balance between consistency and diversity in outputs
|
| 80 |
+
- May exhibit some stylistic tendencies inherited from the training process
|
|
|
|
| 81 |
|
| 82 |
# Training Approach:
|
| 83 |
|
| 84 |
+
- Utilizes the extensive Proteus datasets
|
| 85 |
+
- Focused on maintaining model capabilities while increasing compatibility
|
| 86 |
|
| 87 |
|
| 88 |
# Custom CLIP Integration:
|
| 89 |
|
| 90 |
+
- Incorporation of a specially trained CLIP model
|
| 91 |
+
- Requires a clip skip setting of 2 for optimal performance
|
|
|
|
| 92 |
|
| 93 |
|
| 94 |
# About Prometheus
|