PavonicAI commited on
Update org card with ForgeAI description
Browse files
README.md
CHANGED
|
@@ -1,10 +1,38 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: ForgeAI
|
| 3 |
+
emoji: "🔥"
|
| 4 |
+
colorFrom: purple
|
| 5 |
+
colorTo: red
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# ForgeAI
|
| 11 |
+
|
| 12 |
+
**Making AI models accessible for consumer hardware.**
|
| 13 |
+
|
| 14 |
+
We quantize large AI models, fix compatibility issues, and document everything so the community can benefit.
|
| 15 |
+
|
| 16 |
+
## What We Do
|
| 17 |
+
|
| 18 |
+
- **Quantize large models** to run on GPUs with 16 GB VRAM or less
|
| 19 |
+
- **Fix compatibility issues** with modern PyTorch, transformers and torchtune versions
|
| 20 |
+
- **Provide ComfyUI integration** for creative AI workflows
|
| 21 |
+
- **Document all fixes** with copy-paste code examples
|
| 22 |
+
|
| 23 |
+
## Our Models
|
| 24 |
+
|
| 25 |
+
| Model | What it does | VRAM | Link |
|
| 26 |
+
|---|---|---|---|
|
| 27 |
+
| **HeartMuLa-3B-4bit** | AI music generation from lyrics | 16 GB | [Download](https://huggingface.co/PavonicAI/HeartMuLa-3B-4bit) |
|
| 28 |
+
|
| 29 |
+
## Tech Stack
|
| 30 |
+
|
| 31 |
+
- PyTorch + CUDA
|
| 32 |
+
- bitsandbytes (NF4 / INT8 quantization)
|
| 33 |
+
- Hugging Face Transformers
|
| 34 |
+
- ComfyUI custom nodes
|
| 35 |
+
|
| 36 |
+
## Get In Touch
|
| 37 |
+
|
| 38 |
+
Open an issue or discussion on any of our model repos.
|