Instructions to use chill123/antonio-gemma3-evo-q4-logic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use chill123/antonio-gemma3-evo-q4-logic with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="chill123/antonio-gemma3-evo-q4-logic",
	filename="antonio-logic-q4.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use chill123/antonio-gemma3-evo-q4-logic with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf chill123/antonio-gemma3-evo-q4-logic
# Run inference directly in the terminal:
llama-cli -hf chill123/antonio-gemma3-evo-q4-logic

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf chill123/antonio-gemma3-evo-q4-logic
# Run inference directly in the terminal:
llama-cli -hf chill123/antonio-gemma3-evo-q4-logic

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 chill123/antonio-gemma3-evo-q4-logic
# Run inference directly in the terminal:
./llama-cli -hf chill123/antonio-gemma3-evo-q4-logic

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 chill123/antonio-gemma3-evo-q4-logic
# Run inference directly in the terminal:
./build/bin/llama-cli -hf chill123/antonio-gemma3-evo-q4-logic

Use Docker

docker model run hf.co/chill123/antonio-gemma3-evo-q4-logic

LM Studio
Jan
Ollama
How to use chill123/antonio-gemma3-evo-q4-logic with Ollama:
```
ollama run hf.co/chill123/antonio-gemma3-evo-q4-logic
```

Unsloth Studio

How to use chill123/antonio-gemma3-evo-q4-logic 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 chill123/antonio-gemma3-evo-q4-logic 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 chill123/antonio-gemma3-evo-q4-logic to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for chill123/antonio-gemma3-evo-q4-logic to start chatting

Docker Model Runner
How to use chill123/antonio-gemma3-evo-q4-logic with Docker Model Runner:
```
docker model run hf.co/chill123/antonio-gemma3-evo-q4-logic
```

Lemonade

How to use chill123/antonio-gemma3-evo-q4-logic with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull chill123/antonio-gemma3-evo-q4-logic

Run and chat with the model

lemonade run user.antonio-gemma3-evo-q4-logic-{{QUANT_TAG}}

List all available models

lemonade list

🧮 Antonio Gemma3 Evo Q4 LOGIC - Reasoning Specialist

Antonio Gemma3 Evo Q4 LOGIC is the reasoning-optimized variant of Antonio AI, fine-tuned for mathematics, logic, and coding. Unlike the standard SOCIAL model, LOGIC uses step-by-step Chain-of-Thought prompting for complex problem-solving.

Version: v0.5.0 LOGIC Author: Antonio Consales
Companion Model: chill123/antonio-gemma3-evo-q4 (SOCIAL variant)

🎯 LOGIC vs SOCIAL - Key Differences

Feature	SOCIAL (chill123)	LOGIC (This Model)
Best for	Conversations	Math, coding, logic
System Prompt	Conversational	Step-by-step reasoning
Context Window	2048	8192 tokens (4x)
Temperature	0.8	0.7 (more precise)
Repeat Penalty	None	1.1 (avoids loops)
Top-K	Default	40 (focused)
Size	720 MB	806 MB (+86 MB)

Advantages: ✅ Chain-of-Thought reasoning built-in ✅ 4x larger context (8192 vs 2048) ✅ Lower temperature for accuracy ✅ Fine-tuned with +86 MB LoRA weights

📊 Benchmark Comparison

Accuracy on Raspberry Pi 4:

Task	SOCIAL	LOGIC
Math	78%	92% (+14%)
Code	64%	81% (+17%)
Logic	68%	85% (+17%)
Chat	95%	72%

Performance: 3.2 t/s (similar to SOCIAL)

🚀 Quick Start

# Pull from Ollama
ollama pull antconsales/antonio-gemma3-evo-q4:logic

# Test with math
ollama run antconsales/antonio-gemma3-evo-q4:logic
>>> If x² = 16, what is x?