---
library_name: gguf
pipeline_tag: text-generation
base_model: LiquidAI/LFM2.5-1.2B-Instruct
tags:
- gguf
- on-device
- finance
- financial-intelligence
- ghost-ai
- crypto
- solana
- privacy
- llama.cpp
- fork
quantized_by: Ghost AI Pro
---

# Ghost AI Pro

**On-device financial intelligence — built for Ghost AI on Solana.**

> Ghost AI Pro is part of **Ghost AI**, an on-device privacy-first crypto application running on **Solana**. The model powers private, local AI features inside the Ghost AI app — your prompts, wallet context, and financial data never leave your device.
>
> Ghost AI Pro is a fork of **LiquidAI's LFM 2.5 (1.2B Instruct)**, rebranded and packaged for on-device financial use cases. All underlying model weights, architecture, and capabilities derive from LFM 2.5 — credit and licensing terms remain with the upstream project.

Ghost AI Pro is a compact, quantized language model engineered to run privately on consumer hardware — phones, laptops, edge devices — without sending a single token to the cloud. It is purpose-tuned for the financial domain: portfolio reasoning, market commentary, statement parsing, expense categorization, on-chain activity summarization, and conversational assistance for personal, professional, and crypto-native finance workflows.

For ultra-light deployments (mobile, embedded), see the smaller **[Ghost AI](https://huggingface.co/immortaltatsu/ghost-ai-gguf)** (350M, Q6_K, ~280MB).

## Ghost AI model family

| Tier | Repo | Base | Quantization | Size | Best for |
|---|---|---|---|---|---|
| **Ghost AI** | `immortaltatsu/ghost-ai-gguf` | LFM 2.5 350M | Q6_K | ~280 MB | Mobile, embedded, instant responses |
| **Ghost AI Pro** (this) | `immortaltatsu/ghost-ai-pro-gguf` | LFM 2.5 1.2B Instruct | Q4_K_S | ~700 MB | Desktop, deeper reasoning |

## Why Ghost AI Pro

- **On-device, always private.** Runs locally via `llama.cpp`, `LM Studio`, `Ollama`, or any GGUF-compatible runtime. Your prompts, wallet data, and financial context never leave your device.
- **Built for finance + crypto.** Optimized prompting and behavior around budgeting, markets, accounting concepts, on-chain transactions, and personal-finance reasoning.
- **Native to Ghost AI on Solana.** Powers the local AI surface of the Ghost AI app — a privacy-first crypto experience on the Solana blockchain.
- **Lean footprint.** Q4_K_S quantization — ~700MB on disk, comfortable on 4GB+ RAM devices.
- **Fast inference.** Designed for real-time, low-latency interaction on CPU and consumer GPU.

## Files

| File | Quantization | Size | Notes |
|---|---|---|---|
| `ghost-ai-pro-Q4_K_S.gguf` | Q4_K_S | ~700 MB | Recommended default. Best balance of quality and footprint for on-device use. |

## Usage

### llama.cpp

```bash
./llama-cli -m ghost-ai-pro-Q4_K_S.gguf -p "Summarize this quarter's spending: ..." -n 256
```

### Ollama (Modelfile)

```
FROM ./ghost-ai-pro-Q4_K_S.gguf
PARAMETER temperature 0.7
PARAMETER num_ctx 4096
```

```bash
ollama create ghost-ai-pro -f Modelfile
ollama run ghost-ai-pro
```

### LM Studio

Download the `.gguf` file and load it directly from the LM Studio model browser.

## Intended Use

Ghost AI Pro is intended for:

- Powering the on-device AI layer of the **Ghost AI** crypto app on Solana
- Private wallet, transaction, and portfolio explanations
- Personal finance assistants and budgeting copilots
- On-device document Q&A over statements, invoices, and reports
- Lightweight market, DeFi, and economic commentary generation
- Embedded assistants in fintech and crypto apps where privacy is non-negotiable

## Limitations

Ghost AI Pro is a small model. It is **not** a substitute for licensed financial advice, accounting, or legal counsel, and it is **not** a trading or transaction-signing agent. Verify any numerical reasoning, dates, ticker symbols, token addresses, on-chain claims, and regulatory statements against authoritative sources before acting on them. Always confirm wallet operations through the Ghost AI app's signing flow — never from raw model output.

## About Ghost AI

**Ghost AI** is an on-device, privacy-first crypto app built on **Solana**. It brings local AI directly into the wallet experience — letting users explore transactions, balances, and DeFi context with a model that runs entirely on their own device. Nothing is logged, sent to a server, or shared with a third party.

Ghost AI Pro is the model that powers this on-device intelligence layer.

## Upstream

Ghost AI Pro is a fork of **[LiquidAI/LFM2.5-1.2B-Instruct](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct)**. The GGUF quantizations are sourced from **[unsloth/LFM2.5-1.2B-Instruct-GGUF](https://huggingface.co/unsloth/LFM2.5-1.2B-Instruct-GGUF)**.

## License

Released for research and on-device application use. This model is governed by the upstream **LFM Open License** issued by Liquid AI — please review the original terms before redistribution or commercial deployment.

---

*Ghost AI Pro — intelligence that stays with you.*