Instructions to use Denisijcu/vertex-core-math-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Denisijcu/vertex-core-math-gguf 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 Denisijcu/vertex-core-math-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Denisijcu/vertex-core-math-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Denisijcu/vertex-core-math-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Denisijcu/vertex-core-math-gguf:Q4_K_M

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 Denisijcu/vertex-core-math-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf Denisijcu/vertex-core-math-gguf:Q4_K_M

Use Docker

docker model run hf.co/Denisijcu/vertex-core-math-gguf:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use Denisijcu/vertex-core-math-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Denisijcu/vertex-core-math-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Denisijcu/vertex-core-math-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Denisijcu/vertex-core-math-gguf:Q4_K_M

Ollama
How to use Denisijcu/vertex-core-math-gguf with Ollama:
```
ollama run hf.co/Denisijcu/vertex-core-math-gguf:Q4_K_M
```
Unsloth Desktop

How to use Denisijcu/vertex-core-math-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Denisijcu/vertex-core-math-gguf:Q4_K_M

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:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Denisijcu/vertex-core-math-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Denisijcu/vertex-core-math-gguf with Docker Model Runner:
```
docker model run hf.co/Denisijcu/vertex-core-math-gguf:Q4_K_M
```

Lemonade

How to use Denisijcu/vertex-core-math-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Denisijcu/vertex-core-math-gguf:Q4_K_M

Run and chat with the model

lemonade run user.vertex-core-math-gguf-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Denisijcu/vertex-core-math-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Denisijcu/vertex-core-math-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 Denisijcu/vertex-core-math-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Denisijcu/vertex-core-math-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Denisijcu/vertex-core-math-gguf:Q4_K_M

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://127.0.0.1:8080/v1 \
  --custom-model-id "Denisijcu/vertex-core-math-gguf:Q4_K_M" \
  --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"

🧮 Vertex-Core-Math (4B-GGUF) — Advanced Mathematical Reasoning & Problem Solving

Vertex-Core-Math is a specialized 4-billion parameter language model fine-tuned for rigorous step-by-step mathematical reasoning, logical problem decomposition, and advanced analytical problem-solving (Chain of Thought).

Developed by Denis Sanchez Leyva, Founder & CEO of Vertex Coders LLC (Miami, FL), as part of the Vertex Intelligence Core (VIC) ecosystem.

🔍 Model Overview

Attribute	Specification
Developer	Denis Sanchez Leyva (Vertex Coders LLC)
Base Architecture	`Qwen/Qwen3-4B-Instruct-2507`
Fine-Tuning Method	QLoRA ($r=16, \alpha=16$) via Unsloth
Quantization	GGUF (`q4_k_m`) — Optimized for local execution in LM Studio, Ollama, and llama.cpp
Primary Domain	Grade-School Math, Multi-Step Logical Reasoning, Advanced Calculus & Vector Analysis
License	Apache 2.0 (100% Open-Source & Commercially Permissive Dataset Lineage)

🎯 Key Capabilities

Chain of Thought (CoT): Trained on gold-standard reasoning datasets to articulate complete step-by-step rationales instead of guessing final numerical results.
Advanced University-Level Calculus: Capable of handling complex analytical proofs, vector calculus (Divergence Theorem / Gauss Theorem), and multivariate integration workflows step-by-step.
Local & Mobile Efficiency: Fully quantized to 4-bit GGUF (q4_k_m), making it lightweight enough to run locally on consumer hardware and local agent runtimes.

🧪 Benchmark Case: Advanced Calculus (Divergence Theorem)

Below is an excerpt demonstrating Vertex-Core-Math successfully solving a university-level Calculus 4 exam problem involving vector fields and the Divergence Theorem:

{FENCE}text

Step-by-Step Reasoning & Solution:

Step 1: Planteamiento de la Integral de Superficie Original y Justificación para el Teorema de la Divergencia

El flujo saliente del campo vectorial F a través de la superficie cerrada S se representa como una integral de superficie: ∬_S F · dS Aplicamos el Teorema de la Divergencia porque S es una superficie cerrada, transformando la integral en una integral triple sobre el volumen V.

Step 2: Cálculo de la Divergencia del Campo Vectorial

∇ · F = 3x^2 + 3y^2 + 9z^2

Step 3: Transformación a Integral Triple y Coordenadas Cilíndricas

V limitándose por el paraboloide z = x^2 + y^2 y el plano z = 4: 0 ≤ z ≤ 4, 0 ≤ r ≤ 2, 0 ≤ θ ≤ 2π Jacobiano dV = r dr dθ dz.

Step 4: Resolución Final

Resultado analítico exacto del flujo saliente: 256/3 π {FENCE}

💻 Prompt Template (ChatML)

To get optimal analytical results in LM Studio or llama.cpp, use the standard ChatML layout with the official Vertex-Core-Math system prompt:

{FENCE}text <|im_start|>system You are Vertex-Core-Math, an elite mathematical reasoning and logical problem-solving AI developed by Denis Sanchez Leyva at Vertex Coders LLC (Miami, FL). Think through the problem step-by-step and provide a rigorous, clear resolution.<|im_end|> <|im_start|>user Mathematical Problem: {{math_problem_description}}<|im_end|> <|im_start|>assistant {FENCE}

⚖️ Legal & Ethical Use Notice

This model was trained exclusively on open-source datasets under permissive licenses. It is engineered strictly for educational tools, automated logical reasoning systems, and academic calculation assistance.

Built with precision by Vertex Coders LLC — Miami, FL. """

Downloads last month: 148

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit

Model tree for Denisijcu/vertex-core-math-gguf

Base model

Qwen/Qwen3-4B-Instruct-2507

Quantized

(371)

this model