add README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model:
|
| 4 |
+
- OpenMOSS-Team/MOSS-Image-SFT
|
| 5 |
+
- darask0/MOSS-Image_4-step_adapter_lora
|
| 6 |
+
- Qwen/Qwen3.5-9B
|
| 7 |
+
- black-forest-labs/FLUX.2-small-decoder
|
| 8 |
+
tags:
|
| 9 |
+
- hexanima
|
| 10 |
+
- qualcomm
|
| 11 |
+
- hexagon
|
| 12 |
+
- npu
|
| 13 |
+
- android
|
| 14 |
+
- text-to-image
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
# HexAnima: MOSS-Image pack
|
| 18 |
+
|
| 19 |
+
MOSS-Image 5B (SFT, Q5_K_XL, 4-step rapid-moss adapter) with the Qwen3.5-9B text encoder and the FLUX.2 small VAE decoder, converted for the Hexagon NPU of Snapdragon 8 Gen 1 / 8+ Gen 1 phones.
|
| 20 |
+
The app (https://github.com/ayanoby/hexanima) downloads these files one by one with resume and sha256 verification (`manifest.json`). They are **not usable with other tools**.
|
| 21 |
+
|
| 22 |
+
About 13 GB: text encoder 6.2 GB, DiT 5.7 GB, token embeddings 0.6 GB, glue 0.2 GB, refiner 0.4 GB, VAE contexts about 70 MB per image size.
|
| 23 |
+
|
| 24 |
+
All components are Apache-2.0, see `NOTICE.md` (authors, sources, changes) and `LICENSE`.
|
| 25 |
+
HexAnima is an independent, unofficial project, not affiliated with the authors named in `NOTICE.md`.
|