Instructions to use zmail-tech/ZPT-Titan-1.2B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zmail-tech/ZPT-Titan-1.2B-Instruct with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="zmail-tech/ZPT-Titan-1.2B-Instruct",
	filename="LFM2.5-1.2B-Instruct.Q4_K_M.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 zmail-tech/ZPT-Titan-1.2B-Instruct 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 zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf zmail-tech/ZPT-Titan-1.2B-Instruct: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 zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf zmail-tech/ZPT-Titan-1.2B-Instruct: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 zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M

Use Docker

docker model run hf.co/zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M

LM Studio
Jan
Ollama
How to use zmail-tech/ZPT-Titan-1.2B-Instruct with Ollama:
```
ollama run hf.co/zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M
```

Unsloth Studio

How to use zmail-tech/ZPT-Titan-1.2B-Instruct 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 zmail-tech/ZPT-Titan-1.2B-Instruct 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 zmail-tech/ZPT-Titan-1.2B-Instruct to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for zmail-tech/ZPT-Titan-1.2B-Instruct to start chatting

How to use zmail-tech/ZPT-Titan-1.2B-Instruct with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf zmail-tech/ZPT-Titan-1.2B-Instruct: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": "zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use zmail-tech/ZPT-Titan-1.2B-Instruct with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf zmail-tech/ZPT-Titan-1.2B-Instruct: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 zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use zmail-tech/ZPT-Titan-1.2B-Instruct with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf zmail-tech/ZPT-Titan-1.2B-Instruct: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 "zmail-tech/ZPT-Titan-1.2B-Instruct: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"

Docker Model Runner
How to use zmail-tech/ZPT-Titan-1.2B-Instruct with Docker Model Runner:
```
docker model run hf.co/zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M
```

Lemonade

How to use zmail-tech/ZPT-Titan-1.2B-Instruct with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull zmail-tech/ZPT-Titan-1.2B-Instruct:Q4_K_M

Run and chat with the model

lemonade run user.ZPT-Titan-1.2B-Instruct-Q4_K_M

List all available models

lemonade list

ZPT-Titan-1.2B-Instruct : Title Generation Model

🚀 Overview

ZPT-Titan-1.2B-Instruct is a highly efficient, 1.2 Billion parameter language model specifically fine-tuned for generating concise and descriptive titles from long-form text.

Built upon the robust LiquidAI LFM 2.5 1.2b architecture and specialized training on the agentlans/wikipedia-paragraphs-titles dataset, this model excels at distilling lengthy content into impactful, accurate, and easily readable titles. It is optimized for integration into applications requiring high-quality content indexing, such as chat interfaces and note-taking tools like Open WebUI.

🛠️ Technical Specifications

This model is provided in optimized formats for broad compatibility and performance.

Feature	Specification	Details
Model Name	ZPT-Titan-1.2B-Instruct	"Titan" references its specialization in Title Generation.
Base Model	LiquidAI LFM 2.5 1.2b	The foundational architecture.
Training Data	`agentlans/wikipedia-paragraphs-titles`	Dataset used to fine-tune title extraction skills.
Fine-tuning Framework	Unsloth Studio	Trained 2x faster using the Unsloth optimization techniques.
Model Size	1.2 Billion Parameters	Offers a strong balance of performance and deployment efficiency.
Supported Formats	GGUF, Quantized	Optimized for CPU/GPU inference via `llama.cpp`.

⚙️ Performance & Implementation

The model has been converted and optimized to the GGUF format, allowing for highly efficient local deployment.

Performance Benefits:

Unsloth Optimization: The training process utilized Unsloth, enabling significantly faster training times and optimized model structure.
GGUF Efficiency: Quantized files like the included Q8_0 version ensure excellent performance even on resource-constrained hardware.

Available Model Files: The following highly optimized file is available for immediate use:

LFM2.5-1.2B-Instruct.Q8_0.gguf

💡 Use Cases

The primary function of ZPT-Titan-1.2B-Instruct is to act as an advanced summarization component focused purely on titling.

Chat Interfaces: Automatically generating compelling titles for conversational threads.
Knowledge Management: Integrating into platforms (e.g., Open WebUI) to auto-title long notes and documents.
Content Indexing: Creating standardized, concise metadata for massive repositories of text.

💻 Installation & Usage

The model is designed to work seamlessly with the llama.cpp ecosystem.

Dependencies:

llama.cpp (for CPU/GPU inference)
unsloth (for initial conversion and optimization)

CLI Usage: You can invoke the model using the following command structure. Ensure you have the necessary tool installed.

Text-Only LLMs:

llama-cli -hf zmail-tech/ZPT-Titan-1.2B-Instruct --jinja

Multimodal Models:

llama-mtmd-cli -hf zmail-tech/ZPT-Titan-1.2B-Instruct --jinja

Note: The model is named Titan to reflect its specialized power and scale in Title Generation. For full setup instructions, please refer to the Unsloth AI GitHub resources.

Downloads last month: 71

GGUF

Model size

1B params

Architecture

lfm2

Hardware compatibility

4-bit

8-bit

Inference Providers NEW

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