What if learning Bitcoin didn’t begin with charts or technical documentation…

…but with a mystery?

The Genesis Heist is an experiment by the Zeus Project exploring a simple idea:

Can storytelling teach cryptography, self-custody, and proof-of-work principles better than tutorials?

Instead of reading about blockchain, participants become treasure hunters — navigating a comic book universe where clues, encryption, and wallets converge into a real-world Bitcoin reward.

And yes — the prize is real.

100,000 Satoshis await the first solver.

🧠 Why Build a Comic Around Bitcoin?

Most people encounter Bitcoin through speculation.

Very few encounter it through understanding.

The Genesis Heist flips onboarding upside down.

Rather than explaining concepts directly, it introduces them naturally through interaction:

• wallets become tools
• cryptography becomes a puzzle
• blockchain becomes a map
• Bitcoin becomes the treasure

Readers don’t just learn Web3.

They use it.

🧩 The Hunt Begins — Searching the Comic Pages

Each page of The Genesis Heist contains subtle clues embedded within artwork, dialogue, and environmental details.

🎲Some clues are visual.
🀄 Others are symbolic.
A few reference cryptographic ideas familiar to Bitcoin users.

Participants must:

1. Explore the comic carefully.
2. Collect fragments of information.
3. Interpret hidden signals.
4. Assemble meaning across multiple pages.

Eventually, hunters discover references pointing toward a cipher.

But the cipher itself is not immediately visible.

🔐 Retrieving the Cipher from the Zeus Community

Treasure hunters must take the next step beyond passive reading.

Inside the Zeus Project Community Telegram group, participants can retrieve a special encoded cipher connected to the story.

This step serves an important purpose:

— encourages community collaboration
— mirrors open-source discovery culture
 — transforms readers into participants

The Telegram group acts like a digital expedition camp — where explorers compare theories and share insights without revealing final solutions.

The cipher is the bridge between fiction and cryptography.

🔍Decrypting the Cipher with Zeus Encryption (Free to Decrypt Tool)

Once obtained, the cipher cannot be understood directly.

It must be decrypted.

To do this, participants use Zeus Encryption, a free to decrypt application built to demonstrate practical cryptographic principles.

The tool uses:

• AES encryption — widely used symmetric encryption standard
• SHA-256 hashing — the same cryptographic foundation used by Bitcoin mining

Rather than reading about encryption academically, users experience:

✅ encoding
✅ decoding
✅ verification
✅ cryptographic reasoning

The act of decrypting becomes part of the story itself.

Technology becomes narrative.

🧭 Why This Matters

The Genesis Heist isn’t just a game.

It explores a deeper hypothesis:

Education in decentralized systems works best when discovery replaces instruction.

Traditional onboarding asks users to trust systems they barely understand.

The Genesis Heist asks users to:

• configure a wallet,
• interact with encryption,
• verify data independently,
• and ultimately claim Bitcoin through knowledge.

This mirrors Bitcoin’s own philosophy:

Don’t trust. Verify.

₿ The Bitcoin Reward — 100,000 Satoshis 💰

Why reward Bitcoin instead of a project token?

Because Bitcoin represents neutral, proof-of-work finality.

Using BTC as the prize:

• anchors value outside the application
• reinforces Bitcoin as monetary settlement
• aligns incentives with decentralized security

The ecosystem becomes educational infrastructure — not monetary competition.

Bitcoin remains the destination.

🔗 Where ZEUS Fits Into the Experience

ZEUS acts as the interaction layer powering the ecosystem:

• wallet identity
• application access
• experimentation tokens
• transaction learning

Participants can safely explore Web3 mechanics using ZEUS before progressing toward deeper self-custody practices.

It is a learning environment built on Proof-of-Work rails via Ethereum Classic.

🌐 Philosophy → Tutorial → Application

The Zeus Project ecosystem now forms three connected layers:

Philosophy
Why blockchain matters in the AI age.

Tutorial
How to configure secure self-custody wallets.

Application
The Genesis Heist — learning through participation.

Together, they demonstrate that decentralized systems are not abstract theories.

They are lived experiences.

🗺️ A New Kind of Onboarding

Bitcoin adoption has historically relied on:

• technical documentation
• financial incentives
• ideological arguments

The Genesis Heist proposes a fourth path:

curiosity-driven discovery.

When users solve puzzles, decrypt messages, and claim rewards themselves, understanding becomes permanent.

🕵️‍♂️ Join the Hunt for the Bitcoin Keys
If you want to participate:

1️⃣ Configure your ZEUS wallet
2️⃣ Read The Genesis Heist comic
3️⃣ Hunt for hidden clues
4️⃣ Retrieve the cipher from the Zeus Telegram community
5️⃣ Use Zeus Encryption to decrypt it
6️⃣ Follow the trail toward Bitcoin

Somewhere inside the story lies the key🔑

Final Thought

Bitcoin secured value through mathematics.

The next challenge is teaching humanity how to interact with that system safely.

Perhaps the best teacher isn’t a manual.

Perhaps it’s a mystery waiting to be solved.

Welcome to The Genesis Heist.

👉Subscribe for : AI Agents, Bitcoin & Cryptographic Identity

Thanks for your time reading this article. If you like the content feel free to follow, clap and share the content on social media. Follow the project:

🔗X: https://x.com/ZeusPayETC 🌐GitHub:https://github.com/ZeusPayETC

📬Telegram: https://t.me/zeusprojectgroup for project updates

🌴LinkTree: https://linktr.ee/Zeus_Project for new project links

A practical guide to self-custody, AI-era security, and powering decentralized applications using MetaMask and Trezor hardware wallets.

The internet is entering a new phase — one where artificial intelligence can act, transact, and make decisions autonomously.

But intelligence without ownership creates risk.

This guide shows how to configure a ZEUS Wallet using MetaMask and Trezor hardware security — giving you sovereign control over assets, identity, and encrypted applications in the emerging decentralized AI economy.

In an age where artificial intelligence is advancing faster than ever, one question becomes unavoidable:

Who controls your digital assets — you, or the system?

As AI grows capable of automating financial systems, managing identities, and interacting autonomously online, the importance of trustless blockchain infrastructure becomes clearer. The future isn’t just intelligent — it must also be secure, verifiable, and self-sovereign.

This is where the ZEUS Wallet enters the picture.

Built on proof-of-work blockchain principles and designed to power a growing ecosystem of decentralized applications, ZEUS enables users to participate in a new digital economy — one where ownership and security are mathematically enforced rather than institutionally promised.

In this guide, we’ll explore:

• Why blockchain matters in the AI era
• How to configure ZEUS Wallet using MetaMask
• How to achieve true self-custody with Trezor hardware wallets
• How ZEUS powers real applications inside the Zeus Project ecosystem
• How to acquire ZEUS via decentralized exchange using ETC

🤖 Blockchain Systems in the Age of Advanced AI

Artificial Intelligence is rapidly becoming agentic — capable of acting, transacting, and managing digital resources independently.

But AI introduces a paradox:

The more powerful automation becomes, the more dangerous centralized control becomes.

Traditional systems rely on databases that can be:

• hacked,
• altered,
• censored,
• or manipulated internally.

Blockchain systems solve this by introducing cryptographic truth.

Proof-of-Work networks like Ethereum Classic and Bitcoin provide:

✅ Immutable transaction history
✅ Decentralized verification
✅ Cryptographic ownership
✅ No central attack point

Unlike centralized platforms, blockchain applications are extremely resistant to hacking because attackers must compromise an entire distributed network rather than a single server.

In the AI age, blockchain becomes the trust layer — ensuring autonomous systems cannot rewrite ownership or falsify transactions.

ZEUS is built precisely for this future.

⚡What is the ZEUS Wallet?

The ZEUS Wallet is not a standalone app — it is a configuration of your Web3 wallet enabling interaction with the Zeus Project ecosystem.

It allows you to:

• Hold ZEUS tokens
• Access decentralized applications
• Participate in puzzles and rewards
• Encrypt and secure data
• Test Web3 applications safely

You can operate ZEUS using:

• 🦊 MetaMask (software wallet)
• 🔐 Trezor Hardware Wallet (cold storage self-custody)

🦊 Step 1 — Configuring ZEUS Wallet on MetaMask
MetaMask acts as your gateway into Web3.

Install MetaMask

1. Install MetaMask extension (Chrome/Firefox/Brave).
2. Create a wallet.
3. Securely store your recovery phrase offline.
4. Import your seed phrase into mobile app to configure the same wallet as the web browser extension version 📲

Always perform strict seed phrase backups on paper and store in secure location. Never expose your seed phrase to the internet or email it to yourself as AI can read all texts and images on the internet!🚨

Add Ethereum Classic Network

ZEUS operates within the Ethereum Classic (ETC) ecosystem — a proof-of-work chain aligned with immutability principles.

Add ETC network manually:

Network Name: Ethereum Classic
RPC URL: https://etc.rivet.link
Chain ID: 61
Currency Symbol: ETC
Block Explorer: https://blockscout.com/etc/mainnet

Import ZEUS Token
After switching to ETC network:

1. Click Import Token
2. Paste ZEUS contract address: 0x66e97838a985cf070b9f955c4025f1c7825de44f
3. Confirm

Your ZEUS balance will now appear.

You are officially connected to the Zeus ecosystem 🎉

🔐 Step 2 — Self-Custody with Trezor Hardware Wallet

Software wallets are convenient.

Hardware wallets are sovereign.

The Trezor Safe 5 allows you to hold ZEUS using Ethereum Classic’s native address system while keeping private keys completely offline.

Why Hardware Matters
Even if:
🔻your computer is compromised,
🔻malware exists,
🔻browsers are attacked,

your private keys never leave the device 🔐

Transactions require physical confirmation.

This eliminates entire categories of hacks.

Self-Custody ZEUS with Trezor Security Hardware

1. Connect Trezor device using USB cable.
2. Open Trezor Suite application.
3. Navigate to your ETC wallet and click receive address.
4. Use this address to send ZEUS to Trezor cold-storage.

Because ZEUS follows ETC token standards, your assets remain fully compatible.

You now control your funds with your own private keys — not an exchange.

🎮 Powering the Zeus Project Ecosystem Applications

ZEUS isn’t theoretical.
It powers live blockchain decentralized applications.

🧩 The Genesis Heist — Earn 100,000 Satoshis (Bitcoin)

The Web3 flipbook comic The Genesis Heist transforms storytelling into an interactive blockchain experience.

Hidden within the comic lies a puzzle.
Solve it first and decipher the encrypted text to win:

💰 100,000 Satoshis (Bitcoin)

Your ZEUS wallet acts as:

- identity,
- proof of participation,
- reward destination.

Discover clues hidden in the pages of the book and join the Zeus Project Telegram group to find the cipher text. Entertainment meets cryptography.

🔐 Zeus Encryption — Secure Data with AES & SHA-256

Modern AI systems generate enormous amounts of sensitive data.

Data privacy becomes a huge problem for humans as advanced AI can now read through and analyse text documents in a matter of minutes.

Zeus Encryption allows users to secure information using:

💠AES encryption — military-grade symmetric security
💠SHA-256 hashing — Bitcoin-level cryptographic integrity

Your wallet becomes a cryptographic identity layer — enabling decentralized data protection without trusting centralized servers.

🚰 Zeus Faucet — Learn by Doing

Zeus Faucet serves as a utility token dispatch interface, where users interested in testing decentralized applications connect their wallets to receive free ZEUS tokens to operate Zeus Mail and Zeus Encryption dApps. Claims limited to 50,000 ZEUS per wallet address on the ETC network.

New users can claim:

⚡ 50,000 ZEUS tokens

from the Zeus Faucet to experiment safely.

This allows anyone to:

🟢test transactions
🟢explore dApps
🟢learn Web3 without financial risk

Education through participation — not theory.

🔄 Acquiring ZEUS via Hebeswap (Decentralized Exchange)

ZEUS can be obtained permissionlessly through Hebeswap, a decentralized exchange on Ethereum Classic.

Process

1. Acquire ETC (Proof-of-Work asset)from a centralized exchange
2. Transfer the ETC to your MetaMask Wallet address
3. Connect MetaMask to Hebeswap. Ensure you’ve added ZEUS from asset list using the ZEUS contract address: 0x66e97838a985cf070b9f955c4025f1c7825de44f
4. Swap ETC → ZEUS ensuring you have enough ETC for gas

No accounts.
No approvals.
No centralized custody.

Just blockchain interaction.

This reinforces a core Zeus philosophy:

Proof-of-Work assets power sovereign digital economies.

🧠 Why Self-Custody Matters More Than Ever

As AI agents begin managing finance, identity, and communication, centralized systems become increasingly fragile.

Self-custody ensures:

✅AI cannot seize your assets.
✅Institutions cannot freeze access.
✅Platforms cannot rewrite ownership.

Your keys = your sovereignty.

Hardware wallets like Trezor combined with decentralized infrastructure create what may be the most resilient digital security model ever invented.

🔮 The Vision Behind Decentralized Applications with Zeus Project 🔮

The Zeus Project explores a future where:

🟢AI operates openly,
🟢encryption protects individuals,
🟢blockchain guarantees truth,
🔐and users remain in control.

ZEUS Wallet is the gateway.

Not just to tokens — but to participation in a decentralized digital civilization.

🚀 Getting Started Today
1️⃣ Install MetaMask
2️⃣ Add Ethereum Classic network
3️⃣ Import ZEUS token
4️⃣ Claim faucet tokens
5️⃣ Secure funds using Trezor
6️⃣ Explore Zeus applications

Final Thoughts on the Evolution of the Internet

The next evolution of the internet will not be defined solely by intelligence — but by trust.

AI may automate the world.

Blockchain ensures it remains free.

And with ZEUS Wallet, that future is already within reach.

⚡ Welcome to Zeus Project Community.

👉Subscribe for Part 2: AI Agents, Bitcoin & Cryptographic Identity

Thanks for your time reading this article. If you like the content feel free to follow, clap and share the content on social media. Follow the project:

🔗X: https://x.com/ZeusPayETC 🌐GitHub:https://github.com/ZeusPayETC

📬Telegram: https://t.me/zeusprojectgroup for project updates

🌴LinkTree: https://linktr.ee/Zeus_Project for new project links

A practical guide to downloading, archiving, and running powerful private open-source AI models offline on your own hardware.

For the first time in computing history, frontier-grade AI models are no longer locked behind cloud APIs.

With open-source releases like GPT-OSS, DeepSeek-R1, Qwen 2.5, and NVIDIA’s PersonaPlex speech-to-speech models, anyone can now own, archive, and run serious AI systems locally — offline, private, and on their own terms.

This article walks you step-by-step through building your own local AI model vault on a 5TB external hard drive, then running those models across:

• A CPU-only laptop
• NVIDIA Jetson edge devices
• A future-facing NVIDIA DGX Spark AI supercomputer

If you can follow a terminal prompt and plug in a hard drive, you can do this.

Why build your own local AI model vault?

There are three big reasons:

1. Sovereignty — “If you don’t own the weights, you don’t own the intelligence.”

2. Longevity — Cloud models change, get rate-limited, or disappear. Your drive doesn’t.

3. Privacy & Cost — Once downloaded, inference is essentially free.

Think of this as cold storage for intelligence.

In this article, we shall explore all the available open-source AI models, their hardware requirements and pros & cons of each hardware.

We shall then explore the nitty gritty details of cloning the LLM repos.

Step 1: Format your 5TB external drive (properly)

Before downloading anything, format the drive correctly. AI models are huge and often exceed tens or hundreds of gigabytes per file.

Recommended file systems

• Windows-first users: NTFS
• Cross-platform (Windows + Linux/macOS): exFAT
• ❌ Avoid FAT32 (file size limits will ruin your day)

Create a clean “LLM Vault” structure

LLM_VAULT/
├── models/
│ ├── gpt-oss/
│ ├── deepseek/
│ ├── qwen/
│ ├── multimodal/
│ └── nvidia/
├── runtimes/
│ ├── llama.cpp/
│ ├── ollama/
│ └── python/
├── cache/
│ └── huggingface/
├── licenses/
├── manifests/
└── notes/

You can let the Python code auto-create these folders in the detailed labs further down this article, just format the drive clean labelled “LLM_VAULT”

“Treat your models like long-term assets, not temporary downloads.”

Step 2: Installing resumable download tools (this is non-negotiable)

Large models will fail mid-download at some point. Resumability matters.

🛠 Core tools 🚧🛠
- Python 3.10+
- Windows PowerShell (Administrator)
- Hugging Face Hub
- Git + Git-LFS (some repos still use it)

Installing Hugging Face tools in PowerShell (admin):

pip install -U huggingface_hub

Set your cache to the external drive so your laptop SSD doesn’t fill up:

export HF_HOME=/path/to/LLM_VAULT/cache/huggingface
export TRANSFORMERS_CACHE=$HF_HOME

“Never let multi-hundred-gigabyte models land on your system drive.” 🚨

Step 3: Download the models (resumable snapshots)
Hugging Face’s snapshot_download gives you:
- Resume on failure
- Full repo integrity
- Exact versions for reproducibility

LLM AI Models Worth Archiving
🔹 GPT-OSS (OpenAI open-weight models)
- gpt-oss-20B — realistic local target
- gpt-oss-120B — archival / DGX-class only

These are Apache-licensed, production-grade models designed for local deployment

🔹 DeepSeek-R1 (reasoning monsters)
- DeepSeek-R1 (full) — enormous, mostly archival
- R1-Distill (1.5B / 7B / 14B / 70B) — the ones you actually run

“If you’re not using a distill, you’re probably not running DeepSeek locally.”

🔹 Qwen 2.5 (text, code, vision, video)
- Qwen2.5-Coder — elite coding performance
- Qwen2.5-VL — vision + video understanding
- Qwen2.5-Omni — text, image, audio, video, streaming speech

🔹 NVIDIA PersonaPlex-7B
- Real-time, full-duplex speech-to-speech
- Designed for conversational agents and assistants

Step 4: Choosing how you’ll run the models
There are three execution lanes. You’ll probably use more than one.

Lane A: llama.cpp (CPU & edge king)
Best for:
- CPU laptops
- NVIDIA Jetson devices
- Quantized GGUF models
- Long-term offline use

“If it runs in llama.cpp, it’ll still run five years from now.”

Lane B: Transformers (GPU & multimodal)
Best for:
- Vision, video, and speech models
- NVIDIA DGX-class machines
- NVIDIA acceleration

Lane C: Ollama (developer convenience)
Best for:
- Quick local chat APIs
- Qwen / DeepSeek text models
- Rapid prototyping

Step 5: What runs where (realistic expectations)
💻 CPU Laptop (16GB RAM)
Works well
✅ Qwen2.5-Coder 1.5B / 7B (quantized)
✅ DeepSeek-R1 Distill 1.5B / 7B
✅ Light GPT-OSS-20B experiments (slow but possible)

Does not work well
🔻 Full DeepSeek-R1 671B
🔻 70B+ models
🔻 Real-time speech or video
“CPU inference trades speed for sovereignty.”

📌 NVIDIA Jetson Nano / Orin (Edge AI)
Jetson Nano (4GB RAM)
- Extremely constrained
- Tiny models only (<1.5B)
- Best used as an AI edge client, not a full LLM host

Jetson Orin (recommended upgrade)
Practical for:
- Quantized 7B models
- Vision pipelines
- Speech capture + inference
- Excellent power-to-performance ratio
“Jetson isn’t about scale — it’s about location.”

🤖 NVIDIA DGX Spark (Local AI supercomputer)
With ~128GB unified memory, DGX Spark unlocks:
💎 DeepSeek-R1 70B class inference
💎 Large Qwen multimodal models
💎 PersonaPlex real-time speech systems
💎 Multi-model orchestration
This is where your archived models come alive.

“DGX Spark turns your model vault into a private AI lab.” 🧪

Hardware Comparison: Pros & Cons
CPU Laptop
✅ Cheapest | Silent | Fully portable | Maximum privacy
🔻 Slow inference | Limited model size | No real-time multimodal

NVIDIA Jetson (Edge)
✅ Low power | Always-on AI | Camera + audio friendly | Great for robotics and gateways
🔻 Memory constrained | Not suited for large LLMs | Requires careful optimization

NVIDIA DGX Spark Supercomputer
✅ Massive memory | Serious AI performance | Runs frontier-scale open models locally | Ideal for research and development
🔻 Expensive | Overkill for casual users | Power and cooling considerations

🔓Detailed Step-By-Step Guide Building LLM AI on 5TB External HDD🔐

Welcome to the age of sovereign intelligence. We are entering an era where AI isn’t just something you use — it’s something you own.

By building your local LLM vault, you protect yourself against platform risk, gain true offline intelligence and future-proof your AI workflows like we do building decentralized applications here at Zeus Project.

This build guide is technically detailed and explicit, carries the benefit of our experience and must be followed meticulously to achieve a working offline LLM vault, so grab a cup of coffee and get comfortable. Let’s dive right in….

We’ll be using Windows PowerShell launched as administrator throughout this guide, prepare some pre-requisites outlined below:

• Install Python 3.10+ : Download the latest Python 64-bit version for Windows from the official website and install it as admin, ensuring you tick the checkbox “Add Python 3.10 to PATH”, “Install launcher for all users” and “Disable path length limit” (if shown).
• Verify Python is Installed Correctly: In PowerShell, run the the check

python --version

If installed correctly, you should see

Python 3.10.x

Also check the pip version using command

pip --version

You should see results like

pip 23.x from ...python310...

You may run into some errors when verifying your Python install such as “Python was not found; run without arguments to install from the Microsoft Store”

This happens because Windows has a fake “python.exe” alias enabled by default. You can resolve this by opening Settings > Apps > Advanced app settings > App execution aliases > Turn of both python.exe and python3.exe and then restart PowerShell.

• Creating a Virtual Environment for your LLM Work: This is highly recommended before you start working on downloading any LLM weights, as this will save you from dependency chaos further down the line. When you launch PowerShell, your terminal usually opens in C:\WINDOWS\system32>. Navigate to your user profile by running the command

PS C:\WINDOWS\system32> cd $env:USERPROFILE

This should navigate your directory to your user profile and then run this command one after the other to create the virtual environment.

PS C:\Users\UserProfile> python -m venv llm-env
PS C:\Users\UserProfile> .\llm-env\Scripts\Activate.ps1

You may get an error “cannot be loaded because running scripts is disabled on this system” because script execution is blocked by default in Windows PowerShell.

🛠 Fix PowerShell Execution Policy: If you encounter this error, follow the fix below:

Step 1: Check Current Policy

Get-ExecutionPolicy -List

You’ll see results below

CurrentUser  Undefined
LocalMachine Restricted

Step 2: Allow script for your user account only:

Execute the command and when prompted to change execution policy, select Y and enter key. Close and restart PowerShell again.

Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

Re-run the command to create the virtual environment:

PS C:\Users\UserProfile> python -m venv llm-env
PS C:\Users\UserProfile> .\llm-env\Scripts\Activate.ps1

If successful, you should see the virtual environment appended to your profile

(llm-env) PS C:\Users\UserProfile>

• Installing and Configuring Hugging Face LLM Repository: If you don’t have an account already, head over to their website huggingface.co and create an account > Settings > Tokens > Click New token > Name it > Set Scope: Read > Copy the token. You will need this when prompted during configuration in your virtual environment.
• Installing Hugging Face Hub Python Packages: Now that we’ve installed Python, configured our virtual environment and created our Hugging Face access tokens, we can now proceed to install the Hugging Face packages using command

python -m pip install -U huggingface_hub

✅ There are a few reasons why this step is crucial as it gives you resumable downloads if they fail, ensures integrity-checked model snapshots and installs Hugging Face CLI tools.

Confirm Hugging Face Hub packages are installed correctly

python -m pip show huggingface-hub

If you see package details, everything is good 🎯

Once the Hugging Face Hub and CLI tools are installed, it’s time to authenticate with the token we created in the previous step.

python -m huggingface_hub login

✅ This is another crucial step in the build process as authenticating allows for higher rate limits, gives access to gated models like NVIDIA PersonaPlex speech-to-speech models and cleaner auth handling.

• Setting your Hugging Face Token as an Environment Variable: We are now ready to authenticate our environment with Hugging Face to allow the download of the LLM weights. Ideally we want this set to be persistent so it can be re-used when we restart PowerShell using this command in order, one after the other.

mkdir $env:USERPROFILE\.huggingface -Force
Set-Content -Path "$env:USERPROFILE\.huggingface\token" -Value "hf_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx" -NoNewline

Replace the -Value “hf_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx” with your own access token generated from Hugging Face inside the quotation marks, and restart PowerShell by closing and re-opening the terminal in admin.

Confirm your token is set successfully with this command

Get-Content "$env:USERPROFILE\.huggingface\token"

💡 You should see your token show up if set correctly.

Run further tests to confirm you’re accessing your account with command

python - << "PY"
import os
from huggingface_hub import whoami, HfApi

token = os.environ.get("HUGGINGFACE_HUB_TOKEN", None)
api = HfApi(token=token)
print("Token env set:", bool(token))
try:
    print("whoami:", api.whoami())
    print("✅ Auth is working")
except Exception as e:
    print("❌ Auth failed:", e)
PY

Your username and account details should now be displayed in terminal✅

💎 Downloading Your LLM AI Weights onto External 5TB HDD 💎

Congratulations getting this far in the pipeline build. In this section, we’ll dive deeper into actually cloning our target repos onto our 5TB external hard drive. We’ll still need to configure our environment to ensure our precious models are archived properly for future inference.

Before we begin, let’s point Hugging Face cache to our 5TB storage target to prevent our local machines from filling up with temp files from extremely large AI tensor files. Perform the commands below one after another, remembering to change the drive letter for your external drive.

setx HF_HOME "E:\LLM_VAULT\cache\huggingface"
setx TRANSFORMERS_CACHE "E:\LLM_VAULT\cache\huggingface"

Close and re-open PowerShell, re-activate your virtual environment to continue.

Verify your cache has been set correctly using the following command

echo $env:HF_HOME

Create a working folder for your Python scripts: This is where we create custom scripts in Python to download the LLM AI repos from Hugging Face. For clarity, the python scripts, notes and tooling will live on your C: drive while your LLM weights are downloaded to your external drive.

mkdir ModelVaultScripts
cd ModelVaultScripts

🧱 Local AI Stack Layout (Recommended Structure)
To keep the system clean, reproducible, and portable, each component of the local AI stack lives in a clearly defined location. This separation is critical when working with very large models and external storage.

| Component    | Location           | Purpose                 | Why this matters                                                            |
|--------------|--------------------|-------------------------|-----------------------------------------------------------------------------|
| Python       | `C:`               | Runtime                 | Keeps the core runtime fast and independent of external drives              |
| Virtual env  | `C:`               | Dependency isolation    | Prevents version conflicts between AI tools and system packages             |
| Scripts      | `ModelVaultScripts`| Control logic           | Allows scripts to be versioned, portable, and reused across machines        |
| Models       | `LLM_VAULT`        | Long-term storage       | Separates massive model files from the OS, making upgrades and backups safe |
| HF cache     | `LLM_VAULT/cache`  | Large temporary files   | Prevents Hugging Face downloads from filling up the system drive            |

This layout mirrors how professional ML teams separate compute, logic, and data, while remaining simple enough to run on a personal machine.

• Run a small “resumable download” test (proves everything works): It’s time to run a small test to confirm all our above configurations are correct, and we can download a small Qwen2.5–0.5B-Instruct LLM AI onto the external drive.

Create a notepad file to hold custom python script with the command

notepad download_test.py

Paste this Python script in the notepad file, replace the drive letter and save with Ctrl + S 💾

import os
from huggingface_hub import snapshot_download

VAULT = r"E:\LLM_VAULT"  # <-- change drive letter if needed

repo = "Qwen/Qwen2.5-0.5B-Instruct"  # small test
out  = os.path.join(VAULT, "models", "test", "Qwen2.5-0.5B-Instruct")

os.makedirs(out, exist_ok=True)

snapshot_download(
    repo_id=repo,
    local_dir=out,
    local_dir_use_symlinks=False,
    resume_download=True
)

print("✅ Downloaded to:", out)

Now run the script using the command and wait some down for Qwen2.5 download to your external drive ⏳

python download_test.py

Verify the files have landed on the external drive either opening the drive manually or checking with this command in PowerShell (replace drive letter to match yours). Check resumables work with Ctrl + C

dir E:\LLM_VAULT\models\test\Qwen2.5-0.5B-Instruct

If successful, you’ll see files like config, tokenizer, model shards, etc

💠 Staged “Big Vault” LLM AI Weights Download ( Safe Order) 💠

Since these LLM model weights are huge, we shall split them into stages to ensure we don’t run into issues. Again, you need to follow these steps meticulously for best results. It is highly recommended you plug in RJ-45 ethernet cable for stable connection.

Staging plan will be elaborated below:

🔵 Stage A (CPU-laptop friendly LLM Models)
 — Qwen2.5-Coder 1.5B / 7B
 — DeepSeek R1 Distill 1.5B / 7B

🟡 Stage B (Multimodal LLM Models)
 — Qwen2.5-VL 7B
 — Qwen2.5-Omni 7B

🟢 Stage C (Archive/DGX LLM Tier)
 — GPT-OSS-20B
 — DeepSeek-R1 671B
 — GPT-OSS-120B

📘 Create Stage A Vault Downloader: Prepare the Python script to download the first stage LLM containing the CPU friendly Qwen2.5-Coder 1.5B/7B and DeepSeek R1 Distill 1.5B/7B. Begin by creating a notepad file

notepad download_vault_stageA.py

Paste the script remembering to change the drive letter to match and save

import os
from huggingface_hub import snapshot_download

VAULT = r"D:\LLM_VAULT"

def dl(repo_id: str, rel_path: str):
    out = os.path.join(VAULT, "models", rel_path)
    os.makedirs(out, exist_ok=True)
    print(f"\n==> Downloading: {repo_id}\n--> To: {out}")
    snapshot_download(
        repo_id=repo_id,
        local_dir=out,
        local_dir_use_symlinks=False,
        resume_download=True,
    )
    print("✅ Done:", repo_id)

# Stage A: CPU-laptop friendly (start here)
MODELS = [
    ("Qwen/Qwen2.5-Coder-1.5B-Instruct", r"qwen\Qwen2.5-Coder-1.5B-Instruct"),
    ("Qwen/Qwen2.5-Coder-7B-Instruct",   r"qwen\Qwen2.5-Coder-7B-Instruct"),
    ("deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", r"deepseek\R1-Distill-Qwen-1.5B"),
    ("deepseek-ai/DeepSeek-R1-Distill-Qwen-7B",   r"deepseek\R1-Distill-Qwen-7B"),
]

for repo, path in MODELS:
    dl(repo, path)

print("\nAll Stage A downloads complete.")

You can now run it and wait for the LLM packages to download

python download_vault_stageA.py

Check the drive to confirm LLM AI packages successfully downloaded 🎉

📒 Create Stage B Vault Downloader: Prepare the Python script to download the second stage LLM containing the multimodal Qwen2.5-VL 7B and Qwen2.5-Omni 7B. Create a notepad file for the script

notepad download_vault_stageB.py

Paste the script remembering to change the drive letter to match and save

import os
from huggingface_hub import snapshot_download

VAULT = r"D:\LLM_VAULT"

def dl(repo_id: str, rel_path: str):
    out = os.path.join(VAULT, "models", rel_path)
    os.makedirs(out, exist_ok=True)
    print(f"\n==> Downloading: {repo_id}\n--> To: {out}")
    snapshot_download(
        repo_id=repo_id,
        local_dir=out,
        local_dir_use_symlinks=False,
        resume_download=True,
    )
    print("✅ Done:", repo_id)

MODELS = [
    ("Qwen/Qwen2.5-VL-7B-Instruct", r"multimodal\Qwen2.5-VL-7B-Instruct"),
    ("Qwen/Qwen2.5-Omni-7B",        r"multimodal\Qwen2.5-Omni-7B"),
]

for repo, path in MODELS:
    dl(repo, path)

print("\nAll Stage B downloads complete.")

You can now run it and wait for the LLM packages to download

python download_vault_stageB.py

Check the drive to confirm LLM AI packages successfully downloaded 🎉

📗 Create Stage C Vault Downloader: Prepare the Python script to download the second stage LLM containing the DGX LLM Tier GPT-OSS-20B, DeepSeek-R1 671B and GPT-OSS-120B. Create a notepad file for the script

notepad download_vault_stageC.py

Because the file sizes here are enormous, we’ll need to make some changes to our environment to prevent Xet token refresh timeouts from Hugging Face. We do this by disabling Xet and setting a higher HTTP timeout in PowerShell.

$env:HF_HUB_DISABLE_XET="1"
$env:HF_HUB_ENABLE_HF_TRANSFER="0"
$env:HF_HUB_HTTP_TIMEOUT="600"

Paste this modified script remembering to change the drive letter to match and save 💾

import os
from huggingface_hub import snapshot_download

VAULT = r"D:\LLM_VAULT"

def dl(repo_id: str, rel_path: str):
    out = os.path.join(VAULT, "models", rel_path)
    os.makedirs(out, exist_ok=True)
    print(f"\n==> Downloading: {repo_id}\n--> To: {out}")
    snapshot_download(
        repo_id=repo_id,
        local_dir=out,
        local_dir_use_symlinks=False,
        resume_download=True,
        max_workers=1, 
    )
    print("✅ Done:", repo_id)

MODELS = [
    ("openai/gpt-oss-20b",  r"gpt-oss\gpt-oss-20b"),
    ("deepseek-ai/DeepSeek-R1", r"deepseek\DeepSeek-R1"),
    ("openai/gpt-oss-120b", r"gpt-oss\gpt-oss-120b"),
]

for repo, path in MODELS:
    dl(repo, path)

print("\nAll Stage C downloads complete.")

You can now run the script and wait for the LLM packages to download

python download_vault_stageC.py

If you struggle with errors downloading GPT-OSS-120B, re-enable Xet using the following command and re-run the python download script again

Remove-Item Env:\HF_HUB_DISABLE_XET -ErrorAction SilentlyContinue

Check the drive to confirm LLM AI packages successfully downloaded.

Congrats on reaching this milestone with a serious offline AI library 🏆

Downloading NVIDIA PersonaPlex- 7B- V1 Speech-to-Speech LLM AI 🗣

Our final download before we proceed to running our CPU friendly models will be the new NVIDIA PersonaPlex-7B-V1, released in 2026 capable of full duplex real-time conversational speech using advanced neural codecs to produce spoken responses. Needless to say this model requires higher powered GPU heavy systems with A100 or H100 NVIDIA chips to run, for customer service research and development purposes. We shall walk through the download process to build the LLM tensor files on external drive.

Since NVIDIA models are gated, you will need to go to Hugging Face UI and accept the terms to gain access to download this open-source model.

Run

python -c "import os; from huggingface_hub import snapshot_download; VAULT=r'D:\LLM_VAULT'; repo='nvidia/personaplex-7b-v1'; out=os.path.join(VAULT,'models','nvidia','personaplex-7b-v1'); os.makedirs(out, exist_ok=True); snapshot_download(repo_id=repo, local_dir=out, local_dir_use_symlinks=False, resume_download=True, max_workers=1); print('✅ PersonaPlex downloaded to:', out)"

PersonaPlex is not a simple “run a single model” setup like Qwen. It’s a pipeline.

“PersonaPlex is a system, not just a model.”

PersonaPlex-7B is a real-time conversational speech system made of multiple components:

Mic Audio
   ↓
Speech-to-Text (ASR)
   ↓
LLM (PersonaPlex-7B)
   ↓
Text-to-Speech (TTS)
   ↓
Speaker Audio

To run it, you’ll need PyTorch, NVIDIA NeMo, TensorRT / TensorRT-LLM and CUDA. “PersonaPlex is a CUDA pipeline, not a CLI toy.”

👨‍💻Running CPU AI Inference on 16GB Laptop using llama.cpp + GGUF 🔮

Well done for getting this far, you now hold some serious open-source AI models in your arsenal, completely offline and in your control.

I’m sure by now everyone is aware of the enormous energy and GPU requirements to train and run LLM AI models, causing companies like OpenAI, xAI, Anthropic and Google AI to invest several billion dollars into data centres that serve AI inference API interfaces, as well as the upcoming multi-billion AI Project Stargate driving up demand for RAM, GPU and SSD storage systems.

Luckily, we have a mechanism for running CPU AI inference on a cheaper 16GB computer, using llama.cpp and GGUF variants of LLM AI weights.

🧠 What Does GGUF Mean in the Context of LLMs?
GGUF stands for:

GGML Unified Format

It is a binary file format designed specifically for running large language models efficiently on local machines, especially CPUs.

GGUF was created as the successor to earlier formats like GGML and is now the primary format used by llama.cpp and similar local inference engines

🔍 What Problem Does GGUF Solve?
Traditional LLMs are usually released in Transformer format (PyTorch or Safe tensors). These are great for:

✅ Training | fine-tuning | GPU inference | research

But they are not optimized for:

🔻 CPU-only machines | laptops with limited RAM | offline use | fast startup times

GGUF exists to bridge that gap.

“GGUF is optimized for inference, not training.”

GGUF files are so portable they contain everything needed to run the model including model weights, tokenizer data, architecture metadata and quantization information.

“Quantization is why a 14GB model can run in under 8GB of RAM.”

Now that you understand the technical differences between transformer weights which we already have on our external hard drives and GGUF models, let’s proceed and clone some models we can run on our laptops:

🦙 Installing llama.cpp into Your Vault Runtime Folder: We’ll now create our target folder on our external drive in runtimes and store a prebuilt binary of llama.cpp from their GitHub release pages using this command:

D:\LLM_VAULT\runtimes\llama.cpp\bin\

Ensure you’ve downloaded the 64-bit Windows package and extract the zip files into the \bin folder. Inspect to confirm you can see llama-cli.exe

Creating Target Folders for GGUF Models in your Vault: We’ll now create our CPU-run models in their separate folders from the transformer models we’ve already archived using the following commands in order:

mkdir D:\LLM_VAULT\models\gguf -Force
mkdir D:\LLM_VAULT\models\gguf\qwen -Force
mkdir D:\LLM_VAULT\models\gguf\deepseek -Force

💎 Downloading Resumable GGUF Qwen & DeepSeek LLM AI: There are several variations of GGUF models in community repos you can run on a 16GB RAM laptop, however I have selected the best quantization in the sweet spot for performance and speed of inference on a CPU below:
✅ qwen2.5-coder-7b-instruct-q4_k_m.gguf (best balance of quality vs RAM/speed)
✅ qwen2.5-coder-7b-instruct-q5_k_m.gguf (better quality, slightly heavier)

Step 1- Download Qwen2.5-Coder-7B-Instruct-Q4_K_M into your vault: Run this Python command remembering to change your drive letter and wait.

python -c "import os; from huggingface_hub import hf_hub_download; VAULT=r'D:\LLM_VAULT'; repo='Qwen/Qwen2.5-Coder-7B-Instruct-GGUF'; filename='qwen2.5-coder-7b-instruct-q4_k_m.gguf'; out=os.path.join(VAULT,'models','gguf','qwen'); os.makedirs(out, exist_ok=True); p=hf_hub_download(repo_id=repo, filename=filename, local_dir=out, local_dir_use_symlinks=False); print('✅ Saved:', p)"

Verify the download completed successfully:

dir D:\LLM_VAULT\models\gguf\qwen

Step 2- Download Qwen2.5-Coder-7B-Instruct-Q5_K_M into your vault:
Run this Python command remembering to change your drive letter and wait.

python -c "import os; from huggingface_hub import hf_hub_download; VAULT=r'D:\LLM_VAULT'; repo='Qwen/Qwen2.5-Coder-7B-Instruct-GGUF'; filename='qwen2.5-coder-7b-instruct-q5_k_m.gguf'; out=os.path.join(VAULT,'models','gguf','qwen'); os.makedirs(out, exist_ok=True); p=hf_hub_download(repo_id=repo, filename=filename, local_dir=out, local_dir_use_symlinks=False); print('✅ Saved:', p)"

Step 3- Running Qwen2.5-Coder-7B-Instruct-Q4_K_M on CPU 👨‍💻:
Double check \bin folder to confirm you can see the file llama-cli.exeYou may see it as just an application extension but confirm you can see the file before proceeding. Run the following command in a block:

D:\LLM_VAULT\runtimes\llama.cpp\bin\llama-cli.exe `
  -m D:\LLM_VAULT\models\gguf\qwen\qwen2.5-coder-7b-instruct-q4_k_m.gguf `
  -t 8 -n 256 `
  -p "You are a helpful coding assistant. Explain Terraform state locking clearly."

You can ask any questions to the LLM from this point for inference.

🐋 Downloading DeepSeek R1 GGUF AI Models:
As observed in previous steps, DeepSeek R1 is a heavy muitl-gigabyte transformer engine that requires a lot of GPU power to run, however there are several distillations in GGUF format we can run on CPUs.

Follow the instructions explicitly to download and run DeepSeek AI:
✅ neody/DeepSeek-R1-Distill-Qwen-7B-gguf(Balanced quality with Thinking)
✅ unsloth/DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf ( Quality for CPU)

Step 1- Download neody/DeepSeek-R1-Distill-Qwen-7B-ggufinto your vault:
Run this Python command remembering to change your drive letter and wait.

python -c "import os; from huggingface_hub import hf_hub_download; VAULT=r'D:\LLM_VAULT'; repo='neody/DeepSeek-R1-Distill-Qwen-7B-gguf'; filename='q5_k_m.gguf'; out=os.path.join(VAULT,'models','gguf','deepseek'); os.makedirs(out, exist_ok=True); p=hf_hub_download(repo_id=repo, filename=filename, local_dir=out, local_dir_use_symlinks=False); print('✅ Saved:', p)"

Rename the file to something descriptive after download is complete

Rename-Item "D:\LLM_VAULT\models\gguf\deepseek\q5_k_m.gguf" "deepseek-r1-distill-qwen-7b-q5_k_m.gguf"

Step 2- Download unsloth/DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf into your vault:
Run the Python code to download DeepSeek-R1–1.5B and rename it for convenience with the command below

python -c "import os; from huggingface_hub import hf_hub_download; VAULT=r'D:\LLM_VAULT'; repo='unsloth/DeepSeek-R1-Distill-Qwen-1.5B-GGUF'; filename='DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf'; out=os.path.join(VAULT,'models','gguf','deepseek'); os.makedirs(out, exist_ok=True); p=hf_hub_download(repo_id=repo, filename=filename, local_dir=out, local_dir_use_symlinks=False); print('✅ Saved:', p)"

Rename-Item "D:\LLM_VAULT\models\gguf\deepseek\DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf" "deepseek-r1-distill-qwen-1.5b-q8_0.gguf"

Step 3 — Running DeepSeek-R1-Distill-Qwen-7B-Q5_K_M.ggufon CPU 👨‍💻:

D:\LLM_VAULT\runtimes\llama.cpp\bin\llama-cli.exe `
  -m D:\LLM_VAULT\models\gguf\deepseek\deepseek-r1-distill-qwen-7b-q5_k_m.gguf `
  -t 8 -n 256 `
  -p "You are a practical reasoning assistant. Explain Terraform state locking and how to fix a stuck lock. Give step-by-step guidance."

DeepSeek-R1–7B is a thinking model so will likely output it’s train of thought before giving an answer. Type “continue” in PowerShell to expand the scope of answers

⛳ Conclusion & Final Thoughts on the Future of LLM AI Systems ⛳

What started as a practical exercise in running open-source language models locally quickly becomes something more profound: a glimpse into where AI is heading and how individuals can take part in shaping that future.

By building a personal, offline-capable AI vault — using models like Qwen 2.5 Coder, DeepSeek R1 (distilled and full variants), GPT-OSS, and NVIDIA PersonaPlex — we’re no longer passive consumers of cloud-hosted intelligence. We become operators, curators, and eventually architects of intelligent systems.

This matters because the next wave of AI isn’t just bigger models — it’s agentic AI: systems that can reason, plan, act, and collaborate across tools and environments. Local inference, modular runtimes, and portable model formats like GGUF are foundational building blocks for that future. They enable experimentation without friction, privacy without compromise, and resilience without dependency.

From a productivity standpoint, AI is already proving itself as a force multiplier:
writing code faster, reasoning through complex infrastructure problems, assisting with research, design, and automation. The more fluent we become in running and shaping these systems locally, the more leverage we gain — both technically and creatively.

At Zeus Project, this philosophy is being applied directly to real-world decentralised applications. Projects like Zeus Encryption, ZEUS Mail, The Genesis Heist, Zeus Faucet, and Zeus Celestial Arts explore how AI can intersect with cryptography, blockchain, education, storytelling, and digital ownership — without sacrificing openness or user sovereignty. AI here isn’t a black box; it’s a tool, embedded thoughtfully into decentralised systems.

The most important takeaway is this: we are still early. Open-source AI is evolving at extraordinary speed, and the people who take time now to understand the models, formats, runtimes, and hardware will be best positioned to build what comes next.

So take this setup, improve it, break it, extend it. Try new models. Explore agent frameworks. Experiment with multimodal systems. Read the papers. Join the communities. The future of AI won’t be defined by a single company or platform — it will be shaped by those willing to learn, build, and share.

And this is just the beginning.

Thanks for your time reading this article. If you like the content feel free to follow, clap and share the content on social media. Follow the project:

🔗X: https://x.com/ZeusPayETC 🌐GitHub:https://github.com/ZeusPayETC

📬Telegram: https://t.me/zeusprojectgroup for project updates

🌴LinkTree: https://linktr.ee/Zeus_Project for new project links

On September 10, 2022, the Genesis block for ZEUS was created on the Ethereum Classic Blockchain.

Inheriting important technical value properties of Bitcoin and Ethereum Classic, ZEUS Coin operates in a Proof-of-Work network environment as an ERC-20 token, offering complete ownership rights and sovereign grade immutability to its holders and participants.

The ZEUS Project aims to empower anyone to unlock instant digital value securely, while helping build Bitcoin Nodes around the world, by incentivizing active participation of engineers to deploy the Lightning Network infrastructure of the future, powered by ZEUS.

ZEUS Coin Real World Utility Use Cases

As a Proof-of-Work token running on the Ethereum Classic Blockchain, $ZEUS has various foundational use cases besides granting a holder complete asset ownership rights.

The main utility focus points for ZEUS in the real world include the following:

• Building Bitcoin Core Nodes — The ZEUS Foundation along with partnering organisations in telecommunications and higher education, will support engineering teams deploy full Bitcoin Nodes, and Bitcoin Layer-2 Network Scaling Lightning Nodes to support the payment networks of the future.

Participating engineering teams benefit from running their own Full Bitcoin Nodes, implementation support, ability to confirm and update transactions on the Bitcoin Blockchain ledger and $ZEUS rewards.

Bitcoin Core Nodes will be leveraged to unlock electrical energy value from renewable energy projects.

Reward for Proof-of-Work — Partnership with specialist waste recycling projects to incentivise community clean-up programs, unlocking employment value using $ZEUS as Proof-of-Work rewards.

StarLink Satellite Communications Deployment — Zeus Project will deploy advanced StarLink satellite communications networks, with NSA high security firewall for institutions and small towns to power businesses.

Some of the practical innovative use cases include:

• Deployment in rural zones with little or no fibre cabling infrastructure.
  •Connectivity for Blockchain mining, HPC and Ai compute hashing facilities.
  •Deployment in schools, hospitals, public spaces with high security encryption.
  •Installation on hi-speed trains and ships with direct access to satellites for uninterrupted hi-speed internet access.
  •Interconnectivity between Blockchain nodes in remote locations on the planet, for secure data storage, messaging and GPU compute sharing multi-networks.
  •Unlocking trapped energy from remote locations in sustainable solar energy projects, efficient resource management and distribution using Decentralized Physical Infrastructure Networks (DePIN) technology.

Greek Gods Digital Art Project — Blockchain networks present exciting opportunities for ownership and experience of digital art at scale. Besides offering private artificial intelligence installations to businesses, the Zeus Project will launch Greek Gods digital art using Llama 2 LLM Ai on Bitcoin Layer-2 network to showcase Bitcoin Name System (BNS) functionality.

To demonstrate Bitcoin Domain Name functionality, holders of the Gods will receive their own .BTC domain names and airdrop of Bitcoin when they bind their Bitcoin Lightning addresses to their God.BTC domain names.

Lightning Loyalty Programs — Incentivizing education and active participation in Layer-2 Lightning Network research and development, secure digital asset cold storage and seed-phrase backup and recovery, with added benefits of earning Satoshis as rewards for routing transactions around a global network cluster of Bitcoin Nodes.

🌎Request live dApps & Node demo https://bit.ly/BTCApplicationDemo

Celestial Eggs of Olympus — We can’t let the Gods have all the fun so in a still to be selected Blockchain, the project will launch the hotly anticipated celestial eggs of Olympus to complement the Zeus ecosystem. Initial offerings are expected to be physical hi-res framed art pieces, with a percentage of proceeds donated to support critical air ambulance services & planting of Oak Trees that support over 1000 plant and animal species.

Once the digital eggs become available onto the Blockchain, anyone with a wallet can mint and own one. To bring educational fun to software engineers, students and enthusiasts, the project will launch treasure hunt events where participants collaborate to solve puzzles and find the eggs nestled on hidden blockchain nodes in vantage points.

Technical Properties of ZEUS

ZEUS is technically a Smart Contracts program running on the Ethereum Classic Blockchain. As a result, all attributes such as store of value, censorship resistance, reliability, security, and decentralization are inherited from the parent ETC smart contracts network.

Ticker: ZEUS

Token Name: ZeusPay

Total Supply: 1,000,000,000,000 ZEUS

Decimals: 18

Token Type: ERC-20

ZEUS Contract Address: 0x66e97838a985cf070b9f955c4025f1c7825de44f

ZEUS Token Distribution

Tokenomics which refers to the distribution structure of $ZEUS in circulation is elaborated below:

Total Supply: 1 Trillion $ZEUS (1,000,000,000,000)

ZEUS Foundation: 55% of $ZEUS (550,000,000,000) for Institutional partnerships, governance and large-scale Bitcoin Node deployment operations. Tokens reserved in time-lock smart contracts.

ZEUS DeX Liquidity: 20% of $ZEUS (200,000,000,000) for limited release on HebeSwap decentralized exchange in open market. Selected centralized exchanges could draw liquidity when necessary.

ZEUS Marketing: 10% of $ZEUS (100,000,000,000) for promotional drives and public relations.

Giveaways/Influencers: 5% of $ZEUS (50,000,000,000) for promotional giveways to ZEUS community members, Web3 influencers and podcasters.

CeX Listings: 5% of $ZEUS (50,000,000,000) for top tier centralized exchange listings.

Liquidity Campaigns: 2.5% of $ZEUS (25,000,000,000) to drive liquidity by offering package of Bitcoin Node + cold hardware security suite + backup and recovery + 33%* $ZEUS in exchange for investment.

ZEUS Reserves: 2.5% of $ZEUS (25,000,000,000)

How to Configure MetaMask on Ethereum Classic Network

MetaMask is a gateway to Web3 technology allowing users on mobile or browser to interact with software applications called dApps built on Blockchain protocols.

At this point, it’s worth noting the difference between Ethereum Classic ETC and Ethereum ETH which is a fork of ETC and therefore share the same algorithmic programming. Both are EVM compatible which enables them to execute Smart Contracts like the one that generated $ZEUS.

Creating a MetaMask account does not require a user to provide their email address or phone number, just a 12-word seed phrase and a password which should be kept secure, not to be shared with anyone and ideally stored in multiple locations. Skip this step if you already have an ETC MetaMask Wallet.

Download MetaMask on your mobile or install the browser extension supported in Chrome, Firefox, Microsoft Edge and the Brave crypto browser.

Follow the instructions to create a new account and record your 12-word seed phrase carefully.

Make sure you set a very strong complex password to protect your MetaMask account. Unlock wallet.

1.MetaMask Initial Interface

Notice after you have unlocked your MetaMask wallet, you are connected to the Ethereum Mainnet with a US$0 ETH balance. You can tap to copy the address and email it to yourself but DO NOT send any ETC to the address at this point.

Spend some time familiarising yourself with the options tabs including viewing your private key.

2.Selecting a Blockchain Network

As a gateway to Web3, MetaMask allows you to connect to several Blockchains besides the default Ethereum network.

Click on the networks drop down menu to reveal other networks, however in our case we will proceed to create a custom RPC network for Ethereum Classic ETC.

3.Creating a Custom RPC for Ethereum Classic Network ETC

Scroll down on the network options and select Custom RPC to begin setting up the Ethereum Classic network.

RPC stands for “Remote Procedure Call” which is a protocol used by machines to call other remote programs on Blockchain networks.

If you are configuring on mobile, populate the settings as shown in the shot above. Copy settings below;

- Network Name: Ethereum Classic

- New RPC URL: https://etc.rivet.link

- Chain ID: 61

- Currency Symbol (optional): ETC

- Block Explorer URL (optional): https://blockscout.com/etc/mainnet/

Once satisfied you have entered the details correctly, click “Save” and Add ETC network to MetaMask.

4.Connecting, Viewing and Managing Your New MetaMask ETC Wallet

Congratulations! You have now completed the process of configuring ETC network on MetaMask.

Notice Ethereum Classic now appears at the top of the wallet and ETC as the main account. Test sending some ETC to your wallet address from an exchange like LBank and don’t forget to allow 0.01 ETC for transaction fees.

It is always advisable to test transactions with small amounts before moving a large amount of funds.

To add $ZEUS Coin to your MetaMask, simply tap ‘Import Tokens’ > Copy & Paste the smart contract in ‘Token Address’ > Tap ‘Token Symbol’ to populate information > Import

ZEUS Smart Contract Address: 0x66e97838a985cf070b9f955c4025f1c7825de44f

How to Earn/Buy ZEUS on Ethereum Classic Network

Anyone can earn ZEUS from Proof-of-Work or expenditure of time and energy to generate value through social impact projects like environmental waste clean-up.

Anyone can provide liquidity by adding a 50/50 weighting of ZEUS/ETC liquidity pair in a pool, and gain $ZEUS rewards from transaction fees.

ZEUS can also be earned from loyalty reward programs initiated by participants in Bitcoin Node infrastructure deployment projects, and Lightning Payment Network implementation projects.

A very limited amount of $ZEUS can be purchased with ETC on decentralized exchange HebeSwap.

MetaMask gives users the ability to browse and interact with other Blockchains and smart contract programs such as a decentralized exchange.

Tap the 3 lines at the top left corner of your MetaMask > Tap Browser.

Once browser window opens, type app.hebeswap.com to open the DeX.

Connect your MetaMask Wallet > Select a Token > Copy and Paste ZEUS Smart Contract address > Add to HebeSwap.

ZEUS Contract Address: 0x66e97838a985cf070b9f955c4025f1c7825de44f

You are now ready to swap ETC for ZEUS and vice versa on HebeSwap.

Creating An Invoice to Receive $ZEUS

An invoicing system has been embedded into the ZEUS Wallet in MetaMask, empowering anyone to send, receive and request $ZEUS with a payment invoice generated in wallet.

Users who prefer a personal touch, can send $ZEUS on their HENS domain names at YourName.ETC

Tap “Receive” to show your ZEUS deposit address which will be the same as your native ETC address

Tap “Request Payment” for the option to manually set any amount of $ZEUS as an invoice.

Enter your desired amount and Click Next.

Send your invoice as an email to recipient or generate a QR code invoice to receive $ZEUS.

Value Preposition for The ZEUS Foundation

The ZEUS Foundation aims to empower computer science students and engineers around the world to deploy Bitcoin Nodes, and Layer-2 Lightning Network Node infrastructure of the future.

Community impact projects in collaboration with specialist recycling technologies, will provide employment for local community groups, while cleaning up the environment of harmful waste.

Educational institutions will benefit from research and development of Web3 technologies and harness their potential to implement solutions to solve real world problems at scale with Blockchain technology.

To conclude our vision, The ZEUS Foundation aims to promote cold storage hardware security best practices in digital asset management, with complete seed-phrase backup and recovery solutions.

If you find ZEUS Project interesting, and would like to unlock more $ZEUS by contributing to the liquidity pool, please send your kind gestures to this ETC/HENS address:

HENS: ZeusDeployer.ETC

ETC: 0xB73040f22ABEb00b0208580F180eC6A8a71B88a5

Thank you for investing your time with The ZEUS Foundation

Follow us for updates on Twitter @ZeusPayETC

🌎Request live dApps & Node demo https://bit.ly/BTCApplicationDemo

Block Scout ZEUS Token URL: https://blockscout.com/etc/mainnet/token/0x66e97838A985cf070B9F955c4025f1C7825de44F/token-transfers#transfers