OpenAI's CLIP-L@336 Text Encoder
Browse files
README.md
CHANGED
|
@@ -1,3 +1,16 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
base_model:
|
| 4 |
+
- openai/clip-vit-large-patch14-336
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
# This is NOT a fine-tune.
|
| 8 |
+
This is the original OpenAI CLIP ViT-L/14@336 Text Encoder, converted to HuggingFace 'transformers' format.
|
| 9 |
+
All credits to the original authors.
|
| 10 |
+
|
| 11 |
+
## Why?
|
| 12 |
+
- See below (Flux.1-dev, CLIP only guidance, CFG 3.5, Heun).
|
| 13 |
+
- For my fine-tuned KO-CLIP ViT-L/14@336 -> [see here](https://huggingface.co/zer0int/CLIP-KO-ViT-L-14-336-TypoAttack)
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
|