Mistral 7B Instruct V0.3 GPTQ 8b

🚀 Mistral-7B-Instruct-v0.3-GPTQ-8b Model

This is a quantized model of Mistral-7B-Instruct-v0.3 using GPTQ developed by IST Austria. It uses the wikitext dataset and the following configuration:

• 8bit
• Act order: True
• Group size: 128

🚀 Quick Start

✨ Features

This model is a GPTQ-quantized version of Mistral-7B-Instruct-v0.3, which can potentially reduce memory usage and speed up inference while maintaining good performance.

📦 Installation

Install vLLM and run the server:

python -m vllm.entrypoints.openai.api_server --model cortecs/Mistral-7B-Instruct-v0.3-GPTQ-8b

💻 Usage Examples

Basic Usage

Access the model:

curl http://localhost:8000/v1/completions     -H "Content-Type: application/json"     -d ' {
        "model": "cortecs/Mistral-7B-Instruct-v0.3-GPTQ-8b",
        "prompt": "San Francisco is a"
    } '

📚 Documentation

Evaluations

We did not check for data contamination. Evaluation was done using Eval. Harness using limit=1000.

Performance