https://github.com/oscartma/deepseek-multimodel-comparison

A project to compare language models like DeepSeek-V3 and Llama3.1 for tasks such as text generation. Includes a FastAPI-based REST API for real-time inference, Docker support for deployment, and a flexible framework for evaluation and experimentation.
https://github.com/oscartma/deepseek-multimodel-comparison

claude-3 deepseek-v3 gpt-4 llama3 qwen2-5

Last synced: 8 months ago
JSON representation

A project to compare language models like DeepSeek-V3 and Llama3.1 for tasks such as text generation. Includes a FastAPI-based REST API for real-time inference, Docker support for deployment, and a flexible framework for evaluation and experimentation.

• Host: GitHub
• URL: https://github.com/oscartma/deepseek-multimodel-comparison
• Owner: OscarTMa
• License: mit
• Created: 2024-12-30T08:48:48.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-12-30T09:54:50.000Z (9 months ago)
• Last Synced: 2025-02-22T04:14:10.929Z (8 months ago)
• Topics: claude-3, deepseek-v3, gpt-4, llama3, qwen2-5
• Homepage:
• Size: 5.86 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# **DeepSeek Multi-Model Comparison**

## **Description**

This project enables the comparison of multiple language models, such as DeepSeek-V3 and Llama3.1, for tasks like text generation. It features a REST API built with FastAPI for real-time inference and Dockerized deployment for easy scalability.

## **Table of Contents**

- [Description](#description)

- [Project Structure](#project-structure)

- [Installation](#installation)

- [Usage](#usage)

- [License](#license)

## **Project Structure**

```

DeepSeek-MultiModel-Comparison/

├── notebooks/

│   ├── model_comparison.ipynb   # Notebook principal para pruebas y análisis.

├── api/

│   ├── app.py                   # API REST para probar modelos (usando FastAPI o Flask).

│   ├── requirements.txt         # Librerías necesarias para el API.

├── models/

│   ├── load_models.py           # Código para cargar DeepSeek-V3 y otros modelos.

├── datasets/

│   ├── input_data.json          # Datos de entrada para pruebas.

├── results/

│   ├── performance_metrics.csv  # Resultados de métricas.

├── README.md                    # Descripción completa del proyecto.

└── Dockerfile                   # Contenedorización con Docker.

```

## **Installation**

1. **Clone the repository:**

   ```bash

   git clone https://github.com/your-username/DeepSeek-MultiModel-Comparison.git

   cd DeepSeek-MultiModel-Comparison

   ```

2. **Build the Docker container:**

   ```bash

   docker build -t deepseek-api .

   ```

3. **Run the container:**

   ```bash

   docker run -p 8000:8000 deepseek-api

   ```

## **Usage**

1. Access the API at `http://localhost:8000/infer`.

2. Provide the following query parameters:

   - `model_name`: Name of the model (e.g., "deepseek" or "llama").

   - `text`: Input text for generation.

3. Example request using `curl`:

   ```bash

   curl -X GET "http://localhost:8000/infer?model_name=deepseek&text=The+future+of+AI+is"

   ```

## **License**

This project is licensed under the MIT License. See the `LICENSE` file for details.