README.md

---
license: other
tags:
  - gguf
  - llama.cpp
  - text-generation
  - commercial
pipeline_tag: text-generation
library_name: llama.cpp
---

# GPT-OSS-20B (Q8_0 GGUF) — Premium Download License

**A 20-billion parameter, high-performance language model quantized to Q8_0 for maximum efficiency and accuracy in local inference.**

## Why This Model?
GPT-OSS-20B (Q8_0) delivers **enterprise-grade performance** for developers, researchers, and businesses who need **powerful AI capabilities without relying on third-party cloud APIs**.  
With **8-bit quantization**, you get **faster inference** and **lower memory usage** — perfect for running on high-end GPUs or CPU-based servers.

**Ideal for:**
- AI-powered customer support systems
- Private chatbots & virtual assistants
- Academic & research projects
- Autonomous AI agents
- Code generation & automation

---

## Licensing & Pricing
This model requires a **paid license to download**.  
All usage rights are **non-commercial by default** unless otherwise agreed in writing.

**Download License:**  
- **Includes:** Single personal or internal-use copy of the model file  
- **No resale, redistribution, or public hosting allowed**

To purchase a download license, email: **support@belto.world**.

---

## Technical Specifications
- **Architecture:** GPT-OSS-20B
- **Quantization:** Q8_0 (8-bit)
- **Format:** GGUF
- **Size:** ~12GB
- **Optimized for:** `llama.cpp`

---

## Quick Start (CLI)
```bash
# Install Git LFS
git lfs install

# Clone the repository (requires purchase & token access)
git clone https://huggingface.co/mich9999/gpt-oss-20b
cd gpt-oss-20b

# Clone and build llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make && cd ..

# Run a prompt
./llama.cpp/llama-cli -m ./gpt-oss-20b.Q8_0.gguf -p "Write a 200-word sales pitch."