Spaces:
Build error
Build error
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,59 +1,123 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
colorTo: blue
|
| 6 |
sdk: gradio
|
| 7 |
pinned: false
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
-
#
|
| 11 |
|
| 12 |
-
|
| 13 |
|
| 14 |
-
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
-
|
| 17 |
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
---
|
| 21 |
|
| 22 |
-
##
|
| 23 |
|
| 24 |
-
|
| 25 |
-
-
|
| 26 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
| 28 |
---
|
| 29 |
|
| 30 |
-
##
|
| 31 |
|
| 32 |
-
|
| 33 |
|
| 34 |
-
|
| 35 |
-
- Community name
|
| 36 |
-
- Short description
|
| 37 |
|
| 38 |
-
|
| 39 |
|
| 40 |
---
|
| 41 |
|
| 42 |
-
|
| 43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
-
|
| 46 |
|
| 47 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
-
|
| 50 |
|
| 51 |
-
|
| 52 |
-
2. **Exploring Projects** – Try out interactive demos and experiments directly in this Space.
|
| 53 |
-
3. **Contributing Ideas** – Share your suggestions, feedback, or open-source projects if the Space allows contributions.
|
| 54 |
-
4. **Engaging with the Community** – Join discussions in comments or any linked channels (e.g., forums, GitHub, or chat platforms if listed).
|
| 55 |
|
| 56 |
-
|
| 57 |
-
>
|
| 58 |
-
> **UNDER CONTSRUCTION**: We are currently refining our community pages and digital resources. Please check back soon for updates as we continue to build and improve our platform
|
| 59 |
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: TinyModels
|
| 3 |
+
emoji: ⚡
|
| 4 |
+
colorFrom: red
|
| 5 |
colorTo: blue
|
| 6 |
sdk: gradio
|
| 7 |
pinned: false
|
| 8 |
+
license: mit
|
| 9 |
---
|
| 10 |
|
| 11 |
+
# TinyModels 🤏⚡
|
| 12 |
|
| 13 |
+
<div align="center">
|
| 14 |
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
|
| 19 |
|
| 20 |
+
### *Big intelligence. Tiny footprint.*
|
| 21 |
|
| 22 |
+
We build **small, fast, production-ready models** that punch way above their weight.
|
| 23 |
+
No bloat. No nonsense. Just models that work — on your laptop, your phone, your API.
|
| 24 |
+
|
| 25 |
+
</div>
|
| 26 |
+
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
+
## 🧠 Philosophy
|
| 30 |
+
|
| 31 |
+
> Most AI labs race to make models bigger.
|
| 32 |
+
> We race to make them **smaller without losing what matters.**
|
| 33 |
+
|
| 34 |
+
Every model we ship follows three rules:
|
| 35 |
+
|
| 36 |
+
```
|
| 37 |
+
1. Fits on a free GPU → deployable by anyone
|
| 38 |
+
2. Beats models twice its size → efficiency over parameters
|
| 39 |
+
3. Ships with clean docs → actually usable in production
|
| 40 |
+
```
|
| 41 |
+
|
| 42 |
+
---
|
| 43 |
+
|
| 44 |
+
## 📦 Models
|
| 45 |
+
|
| 46 |
+
| Model | Task | Size | Status |
|
| 47 |
+
|---|---|---|---|
|
| 48 |
+
| 🛡️ **CORTYX v2** | 17-label toxicity & AI safety classifier | 141M | ✅ Live |
|
| 49 |
+
| 🔬 **MCLM v1** | 65-label multi-domain classifier | 183M | 🔜 Soon |
|
| 50 |
+
| 📌 **TinyEmbed** | Lightweight sentence embeddings | 32M | 📅 Planned |
|
| 51 |
+
| ⚡ **TinyNER** | Named entity recognition | <50M | 📅 Planned |
|
| 52 |
+
| 💬 **TinyIntent** | Ultra-fast intent classifier | <50M | 📅 Planned |
|
| 53 |
|
| 54 |
---
|
| 55 |
|
| 56 |
+
## ⚡ Why Tiny?
|
| 57 |
|
| 58 |
+
| | Big Models | TinyModels |
|
| 59 |
+
|---|---|---|
|
| 60 |
+
| **Cost** | $$$ per inference | Free / near-free |
|
| 61 |
+
| **Latency** | 500ms–2s | 20–80ms |
|
| 62 |
+
| **Deployment** | Needs A100 | Runs on T4, CPU, phone |
|
| 63 |
+
| **Accessibility** | Labs only | Everyone |
|
| 64 |
+
| **Carbon** | 🔥 High | 🌱 Low |
|
| 65 |
|
| 66 |
---
|
| 67 |
|
| 68 |
+
## 🤝 Open to Everyone
|
| 69 |
|
| 70 |
+
TinyModels is a **free, open community** — anyone can join, contribute models, datasets, or ideas.
|
| 71 |
|
| 72 |
+
No gatekeeping. No paywalls. Just good models, open to all.
|
|
|
|
|
|
|
| 73 |
|
| 74 |
+
> 👉 **Want to contribute?** Follow the org and reach out!
|
| 75 |
|
| 76 |
---
|
| 77 |
|
| 78 |
+
## 🛠️ Built With
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
|
| 83 |
+
|
| 84 |
|
| 85 |
+
Every model is trained on **Google Colab free tier** to prove it doesn't take a supercomputer to build something great.
|
| 86 |
|
| 87 |
+
---
|
| 88 |
+
|
| 89 |
+
## 🗺️ Roadmap
|
| 90 |
+
|
| 91 |
+
```
|
| 92 |
+
✅ CORTYX v2 — 17-label toxicity classifier
|
| 93 |
+
🔜 MCLM v1 — 65-label multi-domain classifier
|
| 94 |
+
📅 CORTYX v3 — improved hate_speech, profanity, self_harm
|
| 95 |
+
📅 TinyEmbed — 32M sentence embeddings
|
| 96 |
+
📅 TinyNER — named entity recognition <50M
|
| 97 |
+
📅 TinyIntent — ultra-fast intent classifier for chatbots
|
| 98 |
+
```
|
| 99 |
+
|
| 100 |
+
---
|
| 101 |
+
|
| 102 |
+
## 🚀 Quick Start
|
| 103 |
+
|
| 104 |
+
All TinyModels work the same way:
|
| 105 |
+
|
| 106 |
+
```bash
|
| 107 |
+
pip install transformers torch sentencepiece huggingface_hub
|
| 108 |
+
```
|
| 109 |
+
|
| 110 |
+
```python
|
| 111 |
+
from huggingface_hub import hf_hub_download
|
| 112 |
+
# Load any model in minutes — see individual model cards for full usage
|
| 113 |
+
```
|
| 114 |
+
|
| 115 |
+
---
|
| 116 |
|
| 117 |
+
<div align="center">
|
| 118 |
|
| 119 |
+
Built with ❤️ by the community · Free forever · Open to all
|
|
|
|
|
|
|
|
|
|
| 120 |
|
| 121 |
+
*Tiny models. Huge ambitions.*
|
|
|
|
|
|
|
| 122 |
|
| 123 |
+
</div>
|