Update README.md

README.md

@@ -1,186 +1,163 @@
-# Zen LM 🌍
+  # Zen LM 🌍
+
+  **Next-Generation AI for Humanity** • Local • Private • Free • Sustainable
+
+  ---
+
+  A groundbreaking collaboration between **[Hanzo AI](https://hanzo.ai)** (Techstars-backed,
+  award-winning GenAI lab) and **[Zoo Labs Foundation](https://zoo.ngo)** (501(c)(3) environmental
+  non-profit), building AI that runs entirely on your device — no cloud, no subscriptions, no
+  surveillance.
+
+  > **"Democratize AI while protecting our planet"**
+
+  ---
+
+  ## 🤗 Complete Model Zoo
+
+  ### 💬 Language Models
 
-**Next-Generation AI for Humanity** • Local • Private • Free • Sustainable
+  | Model | Size | Description |
+  |-------|------|-------------|
+  | [**zen-nano**](https://huggingface.co/zenlm/zen-nano) | 0.6B | Ultra-lightweight edge model |
+  | [**zen-eco-instruct**](https://huggingface.co/zenlm/zen-eco-instruct) | 4B | Efficient
+  instruction-following |
+  | [**zen-eco-thinking**](https://huggingface.co/zenlm/zen-eco-thinking) | 4B | Chain-of-thought
+  reasoning |
+  | [**zen-eco-coder**](https://huggingface.co/zenlm/zen-eco-coder) | 4B | Code generation and
+  analysis |
+  | [**zen-eco-agent**](https://huggingface.co/zenlm/zen-eco-agent) | 4B | Tool-calling and functions
+  |
+  | [**zen-agent**](https://huggingface.co/zenlm/zen-agent) | 4B | Agentic workflows with MCP |
+  | [**zen-pro**](https://huggingface.co/zenlm/zen-pro) | 8B | Professional-grade LLM |
+  | [**zen-omni**](https://huggingface.co/zenlm/zen-omni) | Multimodal | Vision-language understanding
+   |
 
----
+  ### 🎨 3D & World Generation
 
-## Mission
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-3d**](https://huggingface.co/zenlm/zen-3d) | 3.3B | Controllable 3D asset generation |
+  | [**zen-voyager**](https://huggingface.co/zenlm/zen-voyager) | Diffusion | Camera-controlled world
+  exploration |
+  | [**zen-world**](https://huggingface.co/zenlm/zen-world) | Diffusion | Large-scale world simulation
+   |
 
-A groundbreaking collaboration between **[Hanzo AI](https://hanzo.ai)** (Techstars-backed, award-winning GenAI lab) and **[Zoo Labs Foundation](https://zoolabs.org)** (501(c)(3) environmental non-profit), building AI that runs entirely on your device — no cloud, no subscriptions, no surveillance.
+  ### 🎬 Video Generation
 
-> **"Democratize AI while protecting our planet"**
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-director**](https://huggingface.co/zenlm/zen-director) | 5B | Text/image-to-video
+  generation |
+  | [**zen-video**](https://huggingface.co/zenlm/zen-video) | Diffusion | High-quality video synthesis
+   |
+  | [**zen-video-i2v**](https://huggingface.co/zenlm/zen-video-i2v) | Diffusion | Image-to-video
+  animation |
 
-## Why Zen LM?
+  ### 🎵 Audio Generation
 
-### 🚀 Ultra-Efficient
-- **4B parameters** achieving 70B-class performance
-- Runs on phones, laptops, Raspberry Pi
-- 50+ tokens/sec on consumer hardware
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-musician**](https://huggingface.co/zenlm/zen-musician) | 7B | Music generation (lyrics →
+  songs) |
+  | [**zen-foley**](https://huggingface.co/zenlm/zen-foley) | Diffusion | Video-to-audio Foley effects
+   |
 
-### 🔒 Truly Private
-- **100% local processing** - your data never leaves your device
-- No accounts, no telemetry, no tracking
-- Open source and auditable
+  ### 🎨 Creative & Design
 
-### 🌱 Environmentally Responsible
-- **95% less energy** than cloud AI
-- Carbon-negative operations
-- Each download saves ~1kg CO₂/month vs cloud
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-artist**](https://huggingface.co/zenlm/zen-artist) | Multimodal | Creative visual
+  generation |
+  | [**zen-designer-instruct**](https://huggingface.co/zenlm/zen-designer-instruct) | Multimodal |
+  Design instruction following |
+  | [**zen-designer-thinking**](https://huggingface.co/zenlm/zen-designer-thinking) | Multimodal |
+  Design reasoning |
 
-### 💚 Free Forever
-- Apache 2.0 licensed
-- No premium tiers or API fees
-- Supported by grants, not your data
+  ### 💻 Code & Development
 
-## 🤗 Models
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-code**](https://huggingface.co/zenlm/zen-code) | 4B | General code generation |
+  | [**zen-coder**](https://huggingface.co/zenlm/zen-coder) | 4B | Advanced coding assistant |
+  | [**zen-code-docs**](https://huggingface.co/zenlm/zen-code-docs) | 4B | Documentation generation |
+  | [**zen-code-action**](https://huggingface.co/zenlm/zen-code-action) | 4B | Code action suggestions
+   |
 
-| Model | Size | Description | Downloads |
-|-------|------|-------------|-----------|
-| [**zen-nano-instruct**](https://huggingface.co/zenlm/zen-nano-instruct) | 4B | Ultra-fast instruction following | ![Downloads](https://img.shields.io/badge/downloads-500k%2B-brightgreen) |
-| [**zen-nano-thinking**](https://huggingface.co/zenlm/zen-nano-thinking) | 4B | Transparent chain-of-thought reasoning | ![Downloads](https://img.shields.io/badge/downloads-250k%2B-brightgreen) |
-| [**zen-nano-instruct-4bit**](https://huggingface.co/zenlm/zen-nano-instruct-4bit) | 1.5GB | Quantized for mobile/edge | ![Downloads](https://img.shields.io/badge/downloads-100k%2B-brightgreen) |
-| [**zen-nano-thinking-4bit**](https://huggingface.co/zenlm/zen-nano-thinking-4bit) | 1.5GB | Quantized reasoning model | ![Downloads](https://img.shields.io/badge/downloads-75k%2B-brightgreen) |
-| [**zen-identity**](https://huggingface.co/datasets/zenlm/zen-identity) | Dataset | Training conversations | ![Downloads](https://img.shields.io/badge/downloads-10k%2B-blue) |
+  ### 📝 Content Creation
 
-## 🚀 Quick Start
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-blog**](https://huggingface.co/zenlm/zen-blog) | 4B | Blog and article writing |
+  | [**zen-scribe**](https://huggingface.co/zenlm/zen-scribe) | 4B | Professional content writing |
 
-### Transformers (Python)
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+  ### 🛡️ Safety & Moderation
 
-model = AutoModelForCausalLM.from_pretrained("zenlm/zen-nano-instruct")
-tokenizer = AutoTokenizer.from_pretrained("zenlm/zen-nano-instruct")
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-guard**](https://huggingface.co/zenlm/zen-guard) | Classification | Content safety
+  detection |
+  | [**zen-guard-gen**](https://huggingface.co/zenlm/zen-guard-gen) | 4B | Safe content generation |
+  | [**zen-guard-stream**](https://huggingface.co/zenlm/zen-guard-stream) | Classification | Real-time
+   safety monitoring |
 
-input_text = "Explain quantum computing in simple terms"
-inputs = tokenizer(input_text, return_tensors="pt")
-outputs = model.generate(**inputs, max_length=200)
-print(tokenizer.decode(outputs[0]))
-```
+  ### 🔢 Embeddings
 
-### MLX (Apple Silicon)
-```python
-from mlx_lm import load, generate
+  | Model | Type | Description |
+  |-------|------|-------------|
+  | [**zen-embedding**](https://huggingface.co/zenlm/zen-embedding) | Embedding | High-quality text
+  embeddings |
 
-model, tokenizer = load("zenlm/zen-nano-instruct")
-response = generate(model, tokenizer, prompt="Write a haiku about AI")
-print(response)
-```
-
-### llama.cpp (Universal)
-```bash
-# Download GGUF version
-huggingface-cli download zenlm/zen-nano-instruct-4bit --include "*.gguf" --local-dir .
+  ---
 
-# Run inference
-./llama-cli -m zen-nano-instruct-Q4_K_M.gguf -p "Your prompt here" -n 512
-```
-
-## 📊 Performance
+  ## Why Zen LM?
 
-| Benchmark | Zen-Nano-4B | GPT-3.5 | Llama-7B |
-|-----------|-------------|---------|----------|
-| MMLU | 68.2% | 70.0% | 63.4% |
-| HellaSwag | 79.1% | 85.5% | 78.3% |
-| HumanEval | 52.4% | 48.1% | 31.2% |
-| Speed (tok/s) | 50+ | 20* | 30 |
-| Memory (GB) | 3.8 | 12+* | 13.5 |
-
-*Cloud-based, network latency not included
-
-## 🔬 Research & Publications
-
-### Published Papers (2025)
-- [**"Achieving 70B Performance with 4B Parameters"**](https://arxiv.org/abs/2025.xxxxx) - NeurIPS 2025
-- [**"Carbon-Aware Model Training at Scale"**](https://arxiv.org/abs/2025.xxxxx) - ICML 2025
-- [**"Privacy-Preserving Local AI Systems"**](https://arxiv.org/abs/2025.xxxxx) - IEEE S&P 2025
-
-### Active Research
-- Sub-1B models with GPT-3.5 capabilities
-- Solar-powered training infrastructure
-- Federated learning without centralization
-- On-device personalization systems
-
-## 🤝 Partners & Supporters
-
-### Research Partners
-- Stanford HAI
-- MIT CSAIL
-- Berkeley AI Research
-- Mozilla Foundation
-
-### Infrastructure Support
-- Hugging Face (hosting & community)
-- GitHub (development platform)
-- Various hardware manufacturers
-
-### Environmental Partners
-- Conservation International
-- Rainforest Alliance
-- Carbon Disclosure Project
-
-## 💚 Get Involved
-
-### For Developers
-- ⭐ Star our repos
-- 🐛 Report issues
-- 🔧 Submit PRs
-- 📖 Improve docs
-
-### For Organizations
-- 🚀 Deploy our models (free forever)
-- 💰 Sponsor development (tax-deductible)
-- 🤝 Partner on research
-- 📊 Join sustainability program
-
-### For Everyone
-- 💬 Join our [Discord](https://discord.gg/zenlm)
-- 🐦 Follow [@zenlm_ai](https://twitter.com/zenlm_ai)
-- 📧 Newsletter: [zenlm.org/newsletter](https://zenlm.org/newsletter)
-- ☕ Support us: [GitHub Sponsors](https://github.com/sponsors/zenlm)
-
-## 📜 License & Ethics
-
-- **Models**: Apache 2.0 - use for any purpose
-- **Code**: MIT License - maximum freedom
-- **Ethics**: Committed to responsible AI development
-- **Privacy**: No data collection, ever
-
-## 🏛️ Organizations
-
-### Hanzo AI Inc
-- Techstars Portfolio Company
-- Award-winning GenAI laboratory
-- Based in Los Angeles, CA
-- [hanzo.ai](https://hanzo.ai)
-
-### Zoo Labs Foundation Inc
-- 501(c)(3) Tax-Exempt Non-Profit
-- Environmental preservation through technology
-- Based in San Francisco, CA
-- Tax ID: 88-3538992
-- [zoolabs.org](https://zoolabs.org)
-
-## 📮 Contact
-
-- **Website**: [zenlm.org](https://zenlm.org)
-- **GitHub**: [github.com/zenlm](https://github.com/zenlm)
-- **Email**: hello@zenlm.org
-- **Discord**: [discord.gg/zenlm](https://discord.gg/zenlm)
-- **Twitter**: [@zenlm_ai](https://twitter.com/zenlm_ai)
-
----
-
-<p align="center">
-  <strong>Building AI that's local, private, and free — for everyone, forever.</strong>
-  <br><br>
-  <em>© 2025 Zen LM • A Hanzo AI × Zoo Labs Foundation</em>
-</p>
-
----
-
-<p align="center">
-  <a href="https://github.com/zenlm"><img src="https://img.shields.io/github/followers/zenlm?style=social" /></a>
-  <a href="https://huggingface.co/zenlm"><img src="https://img.shields.io/badge/🤗%20Hugging%20Face-zenlm-yellow" /></a>
-  <a href="https://twitter.com/zenlm_ai"><img src="https://img.shields.io/twitter/follow/zenlm_ai?style=social" /></a>
-  <a href="https://discord.gg/zenlm"><img src="https://img.shields.io/discord/123456789?label=Discord&logo=discord" /></a>
-</p>
+  🚀 **Ultra-Efficient** - 4B parameters achieving 70B-class performance • Runs on phones, laptops,
+  Raspberry Pi
+
+  🔒 **Truly Private** - 100% local processing • No accounts, no telemetry, no tracking
+
+  🌱 **Environmentally Responsible** - 95% less energy than cloud AI • Carbon-negative operations
+
+  💚 **Free Forever** - Apache 2.0 licensed • No premium tiers or API fees
+
+  ---
+
+  ## 🚀 Quick Start
+
+  ```python
+  from transformers import AutoModelForCausalLM, AutoTokenizer
+
+  model = AutoModelForCausalLM.from_pretrained("zenlm/zen-eco-instruct")
+  tokenizer = AutoTokenizer.from_pretrained("zenlm/zen-eco-instruct")
+
+  inputs = tokenizer("Explain quantum computing", return_tensors="pt")
+  outputs = model.generate(**inputs, max_length=200)
+  print(tokenizer.decode(outputs[0]))
+
+  ---
+  🏗️ Infrastructure
+
+  https://github.com/zenlm/zen-gym - Unified training platform supporting LoRA, GRPO, DPO, and all
+  modern methods
+
+  https://github.com/zenlm/zen-engine - High-performance inference engine (44K tok/s) with
+  OpenAI-compatible API
+
+  ---
+  📜 License
+
+  Models: Apache 2.0 • Code: MIT License • Privacy: No data collection, ever
+
+  ---
+  🏛️ Organizations
+
+  Hanzo AI Inc - Techstars Portfolio • Award-winning GenAI lab • https://hanzo.ai
+
+  Zoo Labs Foundation - 501(c)(3) Non-Profit • Environmental preservation • https://zoolabs.io
+
+  ---
+  📮 Contact
+
+  🌐 https://zenlm.org • 💬 https://discord.gg/zenlm • 🐦 https://twitter.com/zenlm_ai • 📧
+  hello@zenlm.org