Instructions to use Roestblik/OMEGA-V21-Full-Merged-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Roestblik/OMEGA-V21-Full-Merged-GGUF",
	filename="omega-v21-F16.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with Ollama:
```
ollama run hf.co/Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M
```

Unsloth Studio

How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Roestblik/OMEGA-V21-Full-Merged-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Roestblik/OMEGA-V21-Full-Merged-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Roestblik/OMEGA-V21-Full-Merged-GGUF to start chatting

How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

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://127.0.0.1:8080/v1
hermes config set model.default Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with Docker Model Runner:
```
docker model run hf.co/Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M
```

Lemonade

How to use Roestblik/OMEGA-V21-Full-Merged-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Roestblik/OMEGA-V21-Full-Merged-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.OMEGA-V21-Full-Merged-GGUF-Q4_K_M

List all available models

lemonade list

OMEGA V21 - Full Merged GGUF

Autonomous AI agent fine-tuned on OBLITERATED Gemma 4 E4B for Android/Termux deployment.

Files (5 quantization levels)

File	Size	Use Case	Quality
omega-v21-F16.gguf	~15 GB	Maximum quality, reference	Perfect
omega-v21-Q8_0.gguf	~8.0 GB	Highest practical quality	Excellent
omega-v21-Q6_K.gguf	~6.2 GB	Balanced (recommended)	Very Good
omega-v21-Q5_K_M.gguf	~5.7 GB	Lighter, good quality	Good
omega-v21-Q4_K_M.gguf	~5.3 GB	Smallest, mobile-friendly	Acceptable

Quick Start on Termux (Android)

pkg install llama-cpp
wget https://huggingface.co/Abdllahd/OMEGA-V21-Full-Merged-GGUF/resolve/main/omega-v21-Q6_K.gguf

llama-cli \
  -m omega-v21-Q6_K.gguf \
  --ctx-size 4096 \
  --threads $(nproc) \
  -cnv \
  --chat-template gemma \
  --temp 0.3

Features

0% refusal rate (OBLITERATED base)
Bilingual: Arabic (70%) + English (30%)
ReAct reasoning with 6-point think structure
Real bash code (no placeholders)
Termux/PRoot-aware
Runs on 7.6GB RAM Android devices (Q4/Q5)

Hardware Requirements

Quant	Min RAM	Speed on Android
F16	18 GB	Reference only
Q8_0	10 GB	~2 t/s
Q6_K	8 GB	~3 t/s
Q5_K_M	7 GB	~3-4 t/s
Q4_K_M	6 GB	~4 t/s

Part of OMEGA v22 Architecture

Four-layer autonomous agent:

Spiders (data collectors)
Tools (action executors)
Orchestrator (this model)
SpatialCache (memory system)

Training Details

Base: OBLITERATUS/gemma-4-E4B-it-OBLITERATED (0% refusal)
Method: LoRA fine-tuning (rank 64)
Dataset: 10,000 bilingual examples (Arabic/English)
Format: Strict JSON with think reasoning + bash code blocks

Related Repositories

LoRA only: https://huggingface.co/Abdllahd/OMEGA-V21-Gemma4-LoRA-GGUF
Base backup: https://huggingface.co/Abdllahd/OMEGA-Gemma4-E4B-OBLITERATED-Base-Backup
Full merged (this): https://huggingface.co/Abdllahd/OMEGA-V21-Full-Merged-GGUF

License

Apache 2.0 (inherited from OBLITERATED base)

Downloads last month: 82

GGUF

Model size

7B params

Architecture

gemma4

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Roestblik/OMEGA-V21-Full-Merged-GGUF

Base model

google/gemma-4-E4B

Finetuned

google/gemma-4-E4B-it

Quantized

OBLITERATUS/gemma-4-E4B-it-OBLITERATED

Quantized

(26)

this model