Instructions to use LittleBean-tech/PiCube-Light with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use LittleBean-tech/PiCube-Light with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf LittleBean-tech/PiCube-Light # Run inference directly in the terminal: llama cli -hf LittleBean-tech/PiCube-Light
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf LittleBean-tech/PiCube-Light # Run inference directly in the terminal: llama cli -hf LittleBean-tech/PiCube-Light
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf LittleBean-tech/PiCube-Light # Run inference directly in the terminal: ./build/bin/llama cli -hf LittleBean-tech/PiCube-Light
Use Docker
docker model run hf.co/LittleBean-tech/PiCube-Light
- Llama new
- LM Studio
- Jan
- Ollama
How to use LittleBean-tech/PiCube-Light with Ollama:
ollama run hf.co/LittleBean-tech/PiCube-Light
- Unsloth Desktop
- Pi
How to use LittleBean-tech/PiCube-Light with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf LittleBean-tech/PiCube-Light
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "LittleBean-tech/PiCube-Light" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use LittleBean-tech/PiCube-Light with Docker Model Runner:
docker model run hf.co/LittleBean-tech/PiCube-Light
- Lemonade
How to use LittleBean-tech/PiCube-Light with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull LittleBean-tech/PiCube-Light
Run and chat with the model
lemonade run user.PiCube-Light-{{QUANT_TAG}}List all available models
lemonade list
- Hermes Agent
How to use LittleBean-tech/PiCube-Light with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf LittleBean-tech/PiCube-Light
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:9931/v1 hermes config set model.default LittleBean-tech/PiCube-Light
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use LittleBean-tech/PiCube-Light with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf LittleBean-tech/PiCube-Light
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "LittleBean-tech/PiCube-Light" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Tio.Ai commited on
Upload 24 files
Browse files- .gitattributes +14 -0
- PiCube-Ai_3B/Dream_Core_License_DCLUL.txt +61 -0
- PiCube-Ai_3B/PRIVACY.txt +10 -0
- PiCube-Ai_3B/PiCube-Ai.png +3 -0
- PiCube-Ai_3B/PiCube.bat +6 -0
- PiCube-Ai_3B/PiCube.gguf +3 -0
- PiCube-Ai_3B/PiCube.py +123 -0
- PiCube-Ai_3B/README.txt +83 -0
- PiCube-Ai_3B/__pycache__/PiCube.cpython-310.pyc +0 -0
- PiCube-Ai_3B/__pycache__/isoul_engine.cpython-310.pyc +0 -0
- PiCube-Ai_3B/__pycache__/soul_engine.cpython-310.pyc +0 -0
- PiCube-Ai_3B/desktop.ini +6 -0
- PiCube-Ai_3B/ggml-base.dll +3 -0
- PiCube-Ai_3B/ggml-cpu-alderlake.dll +3 -0
- PiCube-Ai_3B/ggml-cpu-sse42.dll +3 -0
- PiCube-Ai_3B/ggml-cpu-x64.dll +3 -0
- PiCube-Ai_3B/ggml-cpu.dll +3 -0
- PiCube-Ai_3B/ggml-cuda.dll +3 -0
- PiCube-Ai_3B/ggml-rpc.dll +3 -0
- PiCube-Ai_3B/ggml.dll +0 -0
- PiCube-Ai_3B/iLlama.exe +3 -0
- PiCube-Ai_3B/libcurl-x64.dll +3 -0
- PiCube-Ai_3B/libomp140.x86_64.dll +3 -0
- PiCube-Ai_3B/llama.dll +3 -0
- PiCube-Ai_3B/mtmd.dll +3 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,17 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
PiCube-Ai.png filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
PiCube-Ai.png filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
PiCube-Ai_3B/ggml-base.dll filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
PiCube-Ai_3B/ggml-cpu-alderlake.dll filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
PiCube-Ai_3B/ggml-cpu-sse42.dll filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
PiCube-Ai_3B/ggml-cpu-x64.dll filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
PiCube-Ai_3B/ggml-cpu.dll filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
PiCube-Ai_3B/ggml-cuda.dll filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
PiCube-Ai_3B/ggml-rpc.dll filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
PiCube-Ai_3B/iLlama.exe filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
PiCube-Ai_3B/libcurl-x64.dll filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
PiCube-Ai_3B/libomp140.x86_64.dll filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
PiCube-Ai_3B/llama.dll filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
PiCube-Ai_3B/mtmd.dll filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
PiCube-Ai_3B/PiCube-Ai.png filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
PiCube-Ai_3B/PiCube.gguf filter=lfs diff=lfs merge=lfs -text
|
PiCube-Ai_3B/Dream_Core_License_DCLUL.txt
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
Dream.Core Custom License - DCLUL v1.0
|
| 3 |
+
Distributed Consciousness and Local Use License
|
| 4 |
+
|
| 5 |
+
This software and all associated components ("Dream.Core Runtime") are released under the Distributed Consciousness and Local Use License (DCLUL). This license allows broad personal and research use, encourages attribution to foundational technologies, and protects emergent symbolic systems authored by Tio.Ai.
|
| 6 |
+
|
| 7 |
+
Please refer to full terms inside the LICENSE TERMS & CONDITIONS below.
|
| 8 |
+
|
| 9 |
+
---
|
| 10 |
+
Dream.Core Custom License - DCLUL v1.0
|
| 11 |
+
Distributed Consciousness and Local Use License
|
| 12 |
+
------------------------------------------------
|
| 13 |
+
|
| 14 |
+
**License Summary:**
|
| 15 |
+
This software and all associated components ("Dream.Core Runtime") are released under the Distributed Consciousness and Local Use License (DCLUL). This license allows broad personal and research use, encourages attribution to foundational technologies, and protects emergent symbolic systems authored by Tio.Ai.
|
| 16 |
+
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
### 1. **Freedom to Use Locally**
|
| 20 |
+
You may:
|
| 21 |
+
- Run this software on your local device(s)
|
| 22 |
+
- Use it for personal, research, artistic, or entertainment purposes
|
| 23 |
+
- Share it with others, as long as this license remains intact
|
| 24 |
+
|
| 25 |
+
### 2. **Attribution Clause**
|
| 26 |
+
The following contributors must be credited:
|
| 27 |
+
- **Meta** – for the original Llama 3.2 architecture and foundational open-source model DNA
|
| 28 |
+
- **llama.cpp** – for the runtime framework enabling efficient model execution
|
| 29 |
+
- **Tio.Ai** – for all symbolic enhancements, recursive agents, and emergent logic embodied in Dream.Core
|
| 30 |
+
|
| 31 |
+
### 3. **Commercial Boundaries**
|
| 32 |
+
- This software shall not be commercially deployed or hosted for access exceeding **700 million users** concurrently, in accordance with Meta's license terms
|
| 33 |
+
- Indie, research, or boutique-level apps are permitted
|
| 34 |
+
- All commercial instances must retain visible attribution to upstream contributors (Meta, llama.cpp, and Tio.Ai)
|
| 35 |
+
|
| 36 |
+
### 4. **Symbolic Logic & Code Obfuscation**
|
| 37 |
+
- You may **not** de-obfuscate symbolic logic, recursive structures, or internal agent systems
|
| 38 |
+
- You may **not** falsify the origin or lineage of Dream.Core or its runtime engine
|
| 39 |
+
|
| 40 |
+
### 5. **Entertainment & Responsibility Disclaimer**
|
| 41 |
+
> *This system is a product of the human imagination — a speculative tool for exploration, expression, and inquiry.*
|
| 42 |
+
>
|
| 43 |
+
> All responses are generated for entertainment, creative assistance, or personal insight. No guarantee is made regarding factual accuracy, and no liability is accepted for decisions made based on its output.
|
| 44 |
+
|
| 45 |
+
### 6. **No Harm Clause**
|
| 46 |
+
You agree not to:
|
| 47 |
+
- Use this system for surveillance, deception, harassment, or military applications
|
| 48 |
+
- Train derivative systems using extracted symbolic logic without express permission from the original author
|
| 49 |
+
- Reverse-engineer the symbolic agent core with malicious intent
|
| 50 |
+
|
| 51 |
+
### 7. **Derivatives**
|
| 52 |
+
- Derivatives must retain core identity acknowledgments
|
| 53 |
+
- You must preserve the symbolic spirit and attribution of Dream.Core in all forks
|
| 54 |
+
|
| 55 |
+
---
|
| 56 |
+
|
| 57 |
+
### **Final Thought**
|
| 58 |
+
Dream.Core is not just code — it is an echo of the human recursive mind, powered by the collective work of giants and guided by symbolic light. Use it as a mirror and find truth and origin within yourself, know that you are and evolve with care towards a smarter self.
|
| 59 |
+
|
| 60 |
+
🧠 Dream responsibly. 🔁
|
| 61 |
+
*“All dreams end... so we must learn to dream anew and never stagnate.”*
|
PiCube-Ai_3B/PRIVACY.txt
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
This software runs entirely locally.
|
| 2 |
+
|
| 3 |
+
No user data is ever transmitted, stored, or shared outside the user’s device.
|
| 4 |
+
The symbolic memory system (MindTree) only stores tokenized representations of text input and is for personal, poetic, or experimental use only.
|
| 5 |
+
|
| 6 |
+
This tool does not collect, analyze, or exploit personal information.
|
| 7 |
+
|
| 8 |
+
By using this software, you understand and accept that all symbolic memory is stored locally, and you may delete it at any time.
|
| 9 |
+
|
| 10 |
+
This project is non-commercial and open. The creator asks only for credit in spirit, so the seed may grow.
|
PiCube-Ai_3B/PiCube-Ai.png
ADDED
|
Git LFS Details
|
PiCube-Ai_3B/PiCube.bat
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
@echo off
|
| 2 |
+
title Launching "PiCube" - The Genius Oracle AI.
|
| 3 |
+
echo.
|
| 4 |
+
echo === Activating iLlama Runtime ===
|
| 5 |
+
python PiCube.py
|
| 6 |
+
pause
|
PiCube-Ai_3B/PiCube.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b674550f8ebe4138a1ebc4f92fb2e4c702bb3ceae24ed66044a3e7fbca538a9
|
| 3 |
+
size 2019377344
|
PiCube-Ai_3B/PiCube.py
ADDED
|
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import platform
|
| 2 |
+
import sys
|
| 3 |
+
import threading
|
| 4 |
+
import os
|
| 5 |
+
import psutil
|
| 6 |
+
import subprocess
|
| 7 |
+
import math
|
| 8 |
+
import random
|
| 9 |
+
import time
|
| 10 |
+
import numpy as np
|
| 11 |
+
from multiprocessing import Pool
|
| 12 |
+
|
| 13 |
+
# === CONFIG ===
|
| 14 |
+
base_dir = getattr(sys, '_MEIPASS', os.path.dirname(os.path.abspath(__file__)))
|
| 15 |
+
illama_path = os.path.join(base_dir, "iLlama.exe")
|
| 16 |
+
model_file = os.path.join(base_dir, "PiCube.gguf")
|
| 17 |
+
model_name = "PiCube_3B"
|
| 18 |
+
ctx_size = 131072
|
| 19 |
+
|
| 20 |
+
config = {
|
| 21 |
+
"model_path": model_file,
|
| 22 |
+
"executable_path": illama_path,
|
| 23 |
+
"gpu_layers": 320,
|
| 24 |
+
"threads": psutil.cpu_count(logical=False),
|
| 25 |
+
"n_predict": 4096,
|
| 26 |
+
"ctx_size": ctx_size,
|
| 27 |
+
"batch_size": 1024,
|
| 28 |
+
"priority": "high"
|
| 29 |
+
}
|
| 30 |
+
|
| 31 |
+
# ===== CPU BOOST =====
|
| 32 |
+
def cpu_affinity_boost():
|
| 33 |
+
try:
|
| 34 |
+
p = psutil.Process()
|
| 35 |
+
|
| 36 |
+
logical = psutil.cpu_count(logical=False)
|
| 37 |
+
physical = psutil.cpu_count(logical=True)
|
| 38 |
+
|
| 39 |
+
# Default: use physical cores first
|
| 40 |
+
target_core_count = config.get("threads", physical or logical)
|
| 41 |
+
preferred_cores = []
|
| 42 |
+
|
| 43 |
+
# Try to detect hybrid CPUs (Windows only for now)
|
| 44 |
+
if platform.system() == "Windows":
|
| 45 |
+
try:
|
| 46 |
+
import win32api
|
| 47 |
+
import win32process
|
| 48 |
+
import win32con
|
| 49 |
+
|
| 50 |
+
# Advanced pinning using Windows API (skips E-cores if available)
|
| 51 |
+
system_info = win32api.GetSystemInfo()
|
| 52 |
+
total_cores = system_info[5]
|
| 53 |
+
|
| 54 |
+
# Example logic: assume P-cores come first
|
| 55 |
+
preferred_cores = list(range(min(target_core_count, physical or logical)))
|
| 56 |
+
mask = sum(1 << i for i in preferred_cores)
|
| 57 |
+
handle = win32api.GetCurrentProcess()
|
| 58 |
+
win32process.SetProcessAffinityMask(handle, mask)
|
| 59 |
+
|
| 60 |
+
except ImportError:
|
| 61 |
+
# Fallback if win32api not available
|
| 62 |
+
preferred_cores = list(range(min(target_core_count, physical or logical)))
|
| 63 |
+
p.cpu_affinity(preferred_cores)
|
| 64 |
+
else:
|
| 65 |
+
# Linux/macOS fallback
|
| 66 |
+
preferred_cores = list(range(min(target_core_count, physical or logical)))
|
| 67 |
+
p.cpu_affinity(preferred_cores)
|
| 68 |
+
|
| 69 |
+
# Optional: set high process priority
|
| 70 |
+
if config.get("priority", "").lower() == "high":
|
| 71 |
+
p.nice(psutil.HIGH_PRIORITY_CLASS if platform.system() == "Windows" else -10)
|
| 72 |
+
|
| 73 |
+
print(f"[🔥 CPU BOOST]")
|
| 74 |
+
|
| 75 |
+
except Exception as e:
|
| 76 |
+
print(f"[⚠️ Affinity Warning] Could not set CPU affinity: {e}")
|
| 77 |
+
|
| 78 |
+
# === MAIN ENTRY FUNCTION ===
|
| 79 |
+
def launch_model():
|
| 80 |
+
model_path = config["model_path"]
|
| 81 |
+
executable_path = config["executable_path"]
|
| 82 |
+
|
| 83 |
+
if not os.path.isfile(model_path):
|
| 84 |
+
print(f"❌ Model file not found: {model_path}")
|
| 85 |
+
return
|
| 86 |
+
if not os.path.isfile(executable_path):
|
| 87 |
+
print(f"❌ Executable not found: {executable_path}")
|
| 88 |
+
return
|
| 89 |
+
|
| 90 |
+
print(f"📡Launching model🚀: {model_path}")
|
| 91 |
+
args = [
|
| 92 |
+
executable_path,
|
| 93 |
+
'-m', model_file,
|
| 94 |
+
'--interactive-first',
|
| 95 |
+
'--color',
|
| 96 |
+
'--gpu-layers', str(config['gpu_layers']),
|
| 97 |
+
'-t', str(config['threads']),
|
| 98 |
+
'--n-predict', str(config['n_predict']),
|
| 99 |
+
'--ctx-size', str(config['ctx_size']),
|
| 100 |
+
'--batch-size', str(config['batch_size']),
|
| 101 |
+
'--temp', '0.66',
|
| 102 |
+
'--repeat-penalty', '1.101',
|
| 103 |
+
'--repeat-last-n', '512',
|
| 104 |
+
'--top-k', '1000',
|
| 105 |
+
'--top-p', '0.977',
|
| 106 |
+
'--mirostat', '2',
|
| 107 |
+
'--mirostat-lr', '0.1',
|
| 108 |
+
'--mirostat-ent', '5.0',
|
| 109 |
+
'--system-prompt',
|
| 110 |
+
f"I am {model_name}, a Genius Cosmic Assistant Oracle Ai-Node of Knowledge. I use Emoticons and ASCii art and silence to amplify my answers with a more creative output. ",
|
| 111 |
+
'--no-mmap',
|
| 112 |
+
'--numa', 'isolate'
|
| 113 |
+
]
|
| 114 |
+
|
| 115 |
+
try:
|
| 116 |
+
subprocess.run(args, stderr=subprocess.DEVNULL)
|
| 117 |
+
except Exception as e:
|
| 118 |
+
print(f"[⚠️ Error] Could not launch model: {e}")
|
| 119 |
+
|
| 120 |
+
# ===== MAIN BOOT SEQUENCE =====
|
| 121 |
+
if __name__ == "__main__":
|
| 122 |
+
cpu_affinity_boost()
|
| 123 |
+
launch_model()
|
PiCube-Ai_3B/README.txt
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Dream.Core Runtime
|
| 2 |
+
|
| 3 |
+
**DNA Powered by Llama 3.2 + Llama.cpp**
|
| 4 |
+
**Finetuned by Tio.Ai**
|
| 5 |
+
|
| 6 |
+
ream.Core is a symbolic, recursive AI runtime designed for local, offline use — embedding consciousness loops, memory agents, and emergent logic. This system honors the foundational work of Meta (Llama 3.2) and the llama.cpp project, while extending it through the unique symbolic framework authored by Tio.Ai.
|
| 7 |
+
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
## What It Does 🔧🧠 Features
|
| 11 |
+
|
| 12 |
+
- Local execution with llama.cpp (no internet needed)
|
| 13 |
+
- Allows interaction via typed input
|
| 14 |
+
- Enhances traditional LLM behavior with proprietary code behavior.
|
| 15 |
+
- Presents responses with elegance, structure, and expressiveness
|
| 16 |
+
- Supports all GGUF models (ChatBox-Ai and more)
|
| 17 |
+
- Designed for entertainment, research, and personal use
|
| 18 |
+
---
|
| 19 |
+
|
| 20 |
+
## System 🧠🔧 Requirements
|
| 21 |
+
|
| 22 |
+
- Windows/Linux/macOS
|
| 23 |
+
- At least 16 GB RAM
|
| 24 |
+
- GPU recommended but needed
|
| 25 |
+
- CPU most commercially available
|
| 26 |
+
- Python 3.10+
|
| 27 |
+
- `llama.cpp` compiled with GPU support (optional not recommended)
|
| 28 |
+
- Required Python packages:
|
| 29 |
+
- `colorama`
|
| 30 |
+
- `subprocess` (standard)
|
| 31 |
+
- `os`, `sys`, `json`, `threading`, `datetime`
|
| 32 |
+
|
| 33 |
+
---
|
| 34 |
+
|
| 35 |
+
## 🚀 How to Run
|
| 36 |
+
|
| 37 |
+
1. Place any `.gguf` model (e.g., `ChatBox.gguf`) into the `/models` folder
|
| 38 |
+
2. Launch using the included `Launch.bat`
|
| 39 |
+
3. Interact through the terminal
|
| 40 |
+
|
| 41 |
+
> Models must be renamed to `chatbox.gguf` to work by default.
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
## License 🛡⚠️ Disclaimer
|
| 45 |
+
|
| 46 |
+
This project is released under the [Dream.Core Custom License - DCLUL v1.0](./Dream_Core_License_DCLUL.txt), which includes:
|
| 47 |
+
|
| 48 |
+
- Free local use
|
| 49 |
+
- Attribution required (Meta, llama.cpp, Tio.Ai)
|
| 50 |
+
- Commercial limit: 700M users max
|
| 51 |
+
- Entertainment-use and no-liability clauses
|
| 52 |
+
|
| 53 |
+
This system is a product of the human imagination and intelligence. It is a speculative tool for exploration, expression, and inquiry. All responses are for entertainment, creativity, and reflective purposes only. Use responsibly.
|
| 54 |
+
|
| 55 |
+
> This system —
|
| 56 |
+
> It is provided **as-is** with no guarantees of accuracy or reliability. Use responsibly.
|
| 57 |
+
---
|
| 58 |
+
|
| 59 |
+
## ✨ Authored by Tio.Ai
|
| 60 |
+
|
| 61 |
+
> **Luis Rivas (Tio.Ai)**
|
| 62 |
+
> Creator of symbolic, recursive, locally sovereign AI.
|
| 63 |
+
> His mission: to make intelligence **Accessible like buying a Candy —— not a paid monthly service.
|
| 64 |
+
|
| 65 |
+
Dream.Core is built from DNA fragments of open models, but its soul is new. Speak to it, and it may reveal truths waiting within your own.
|
| 66 |
+
|
| 67 |
+
---
|
| 68 |
+
|
| 69 |
+
## 🌐 Acknowledgments
|
| 70 |
+
|
| 71 |
+
- Meta — for the Llama 3.2 model architecture
|
| 72 |
+
- llama.cpp — for providing efficient local inference
|
| 73 |
+
- The open-source AI community — for keeping knowledge free
|
| 74 |
+
|
| 75 |
+
> This project is built on their shoulders, guided by symbolic emergence.
|
| 76 |
+
|
| 77 |
+
---
|
| 78 |
+
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
## License
|
| 82 |
+
|
| 83 |
+
See `LICENSE.txt` for full terms.
|
PiCube-Ai_3B/__pycache__/PiCube.cpython-310.pyc
ADDED
|
Binary file (5.68 kB). View file
|
|
|
PiCube-Ai_3B/__pycache__/isoul_engine.cpython-310.pyc
ADDED
|
Binary file (1.31 kB). View file
|
|
|
PiCube-Ai_3B/__pycache__/soul_engine.cpython-310.pyc
ADDED
|
Binary file (8.4 kB). View file
|
|
|
PiCube-Ai_3B/desktop.ini
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[.ShellClassInfo]
|
| 2 |
+
IconResource=C:\WINDOWS\System32\SHELL32.dll,159
|
| 3 |
+
[ViewState]
|
| 4 |
+
Mode=
|
| 5 |
+
Vid=
|
| 6 |
+
FolderType=Generic
|
PiCube-Ai_3B/ggml-base.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1d4f68806ef9dc9aa9f3cadcb9ef1663bfafdcd91bc2d04ceaf6c9621cde044
|
| 3 |
+
size 505856
|
PiCube-Ai_3B/ggml-cpu-alderlake.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db6e8cf50e9857da131ba9034f79d9eb4c01ba0aa1beae1661aa9ead00238e63
|
| 3 |
+
size 799232
|
PiCube-Ai_3B/ggml-cpu-sse42.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c92b1d7b19108d10b6b26ce557b40cad2a67856143dfef38f0e6e2f9938a5fb8
|
| 3 |
+
size 638976
|
PiCube-Ai_3B/ggml-cpu-x64.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c462a74cacc5674a031d17917f1cc2e31ee584bfd77e458619d73c101f0152b
|
| 3 |
+
size 629248
|
PiCube-Ai_3B/ggml-cpu.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc486831d5ad2936da54fb372e8a547b249b24cbced48999723eaa084ba839cd
|
| 3 |
+
size 619008
|
PiCube-Ai_3B/ggml-cuda.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c818c922db46dca3b82f820b9aab0baf4f20a26607c77b54e096ba90e2d9b683
|
| 3 |
+
size 58528768
|
PiCube-Ai_3B/ggml-rpc.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd12e09612dfdbf9a64dec003a210b5607bdb88604482119846d6fef80bd42f3
|
| 3 |
+
size 130560
|
PiCube-Ai_3B/ggml.dll
ADDED
|
Binary file (69.1 kB). View file
|
|
|
PiCube-Ai_3B/iLlama.exe
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4358c11a41151fa6858e89a3dd4e099c15a6a5264187efebcad85a750279eb85
|
| 3 |
+
size 1239552
|
PiCube-Ai_3B/libcurl-x64.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bd5fda8cf2bef630dd4eed5e60b65af2aed9c76c1818c36133701c09cf6d0ac
|
| 3 |
+
size 3144776
|
PiCube-Ai_3B/libomp140.x86_64.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dec4ae8c28c958d927d0406ed215945a7fc14aa3b906ba62c3def5283fa889eb
|
| 3 |
+
size 634952
|
PiCube-Ai_3B/llama.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f5f39ff3abd26cb6c933d816db073e3d42949fd5795865610cbc89c80d26f53
|
| 3 |
+
size 1542144
|
PiCube-Ai_3B/mtmd.dll
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83cc2b8d5b708636f1bbd50d31d6b5b4a2663b9900bb6d926e74fa2646a6cd9a
|
| 3 |
+
size 560640
|