Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -22,7 +22,9 @@ tags:
|
|
| 22 |
> use case until v1.2. No high-resolution adaptation pass yet: at high
|
| 23 |
> resolutions (especially two-person edits) identities can bleed together —
|
| 24 |
> prefer ~1–1.5MP and upscale. v1 remains available for workflow
|
| 25 |
-
> reproducibility.
|
|
|
|
|
|
|
| 26 |
|
| 27 |
**Instruction-based, identity-preserving image editing for Krea 2** (12.9B
|
| 28 |
single-stream MMDiT). Give it an image and a plain-language instruction; it
|
|
|
|
| 22 |
> use case until v1.2. No high-resolution adaptation pass yet: at high
|
| 23 |
> resolutions (especially two-person edits) identities can bleed together —
|
| 24 |
> prefer ~1–1.5MP and upscale. v1 remains available for workflow
|
| 25 |
+
> reproducibility. **Low-VRAM variants:** `..._v1_1_r128.safetensors` (0.91GB)
|
| 26 |
+
> and `..._v1_1_r64.safetensors` (0.46GB) — SVD rank-reduced from v1.1,
|
| 27 |
+
> near-identical quality.
|
| 28 |
|
| 29 |
**Instruction-based, identity-preserving image editing for Krea 2** (12.9B
|
| 30 |
single-stream MMDiT). Give it an image and a plain-language instruction; it
|