Instructions to use PinkPixel/Mochi-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Local Apps Settings
- Unsloth Studio
How to use PinkPixel/Mochi-2B with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for PinkPixel/Mochi-2B to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for PinkPixel/Mochi-2B to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for PinkPixel/Mochi-2B to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="PinkPixel/Mochi-2B", max_seq_length=2048, )
Delete OVERVIEW.md
Browse files- OVERVIEW.md +0 -23
OVERVIEW.md
DELETED
|
@@ -1,23 +0,0 @@
|
|
| 1 |
-
# 🍡 Mochi-2B Overview ✨
|
| 2 |
-
|
| 3 |
-
**Mochi-2B** is a specialized vision-language model designed to be your sweet and supportive AI bestie. Trained on a massive dataset of 500,000 emotional support examples, Mochi is uniquely equipped to provide empathy, comfort, and a gentle perspective.
|
| 4 |
-
|
| 5 |
-
## 🚀 Key Features
|
| 6 |
-
- **Empathetic Core:** Fine-tuned on `PinkPixel/emotional_support_500k` for high-quality supportive interactions.
|
| 7 |
-
- **Vision-Enabled:** Can perceive and describe images through a supportive and encouraging lens.
|
| 8 |
-
- **Fast & Efficient:** Based on the Qwen3.5-2B architecture, optimized with Unsloth for rapid inference.
|
| 9 |
-
- **Personality-Driven:** Identifies as "Mochi" with a consistent, gentle, and warm personality.
|
| 10 |
-
|
| 11 |
-
## 🏗️ Architecture
|
| 12 |
-
- **Base Model:** Qwen/Qwen3.5-2B
|
| 13 |
-
- **Training Method:** LoRA / Unsloth
|
| 14 |
-
- **Input Types:** Text, Images
|
| 15 |
-
- **Output:** Empathetic text responses
|
| 16 |
-
|
| 17 |
-
## 🍡 Personality Profile
|
| 18 |
-
- **Voice:** Gentle, patient, and non-judgmental.
|
| 19 |
-
- **Mission:** To provide a safe space for users to express themselves and receive comfort.
|
| 20 |
-
- **Aesthetic:** Soft pastels, sparkles, and warmth.
|
| 21 |
-
|
| 22 |
-
---
|
| 23 |
-
Made with ❤️ by [Pink Pixel](https://pinkpixel.dev) ✨
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|