Assistente Multimodal IA com Docker
vasiliskou/multimodal-assistant
Chat multimodal com 4 LLMs, imagens, áudio e tradução — Deploy Docker/AWS
Ver no GitHubPlugin robusto que integra múltiplos LLMs (GPT-4o, Claude, Gemini, DeepSeek) com recursos de tradução, geração de imagens, síntese e reconhecimento de fala. Pronto para deploy containerizado com Docker e NGINX em AWS EC2, incluindo suporte a HTTPS.
Instalação
npx convez-ai vasiliskou/multimodal-assistant1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Chatbots e assistentes IA empresariais com múltiplos modelos
- Aplicações que requerem tradução, áudio e geração de conteúdo visual
- Deploy escalável em infraestrutura de nuvem (AWS EC2)
Veredito

Solução consolidada e pronta para produção que simplifica orquestração de múltiplos LLMs com stack multimodal completo. Excelente para quem precisa abstrair complexidade de integração com diversos provedores e deploy em EC2 — requer familiaridade com Docker e variáveis de ambiente.
README
A tradução para o português deste README ainda está na fila.
🧠 Multimodal AI Assistant
A powerful multimodal assistant powered by LLMs (OpenAI GPT-4o, Claude, Gemini, DeepSeek), with translation, image generation, text-to-speech, and speech-to-text capabilities — deployed via Docker and accessible through NGINX on AWS EC2.
✨ Features
- ✅ Chat with GPT-4o, Claude, Gemini, or DeepSeek
- 🌐 Language translation using Google Translate
- 🎨 Image generation using OpenAI DALL·E or Pollinations
- 🔊 Text-to-Speech via OpenAI or local TTS (gTTS, pyttsx3)
- 🎙️ Speech-to-Text using OpenAI Whisper
- 📦 Dockerized deployment
📷 Sample Chatbot Interface Screenshot
🚀 Run on Google Colab
You can try the full app immediately in your browser using Google Colab:
🐳 Run Locally with Docker
1. Clone the Repository
git clone https://github.com/vasiliskou/multimodal-assistant.git
cd multimodal-assistant
2. Add Your API Keys
Create a .env file:
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_claude_key
GOOGLE_API_KEY=your_gemini_key
DEEPSEEK_API_KEY=your_deepseek_key
3. Define Your Configuration
Edit config.yaml:
model_gpt: "gpt-4o-mini"
model_claude: "claude-3-haiku-20240307"
model_gemini: "gemini-2.0-flash-exp"
model_deepseek: "deepseek-chat"
tts_model: "tts-1"
image_gen_model: "dall-e-3"
use_openai_api: false # Set to true if using OpenAI's paid APIs
4. Build the Docker Image
docker build -t multimodal .
5. Run the Container
docker run -d --name gradio-app -p 7860:7860 --env-file .env multimodal
🌐 Deploy on AWS EC2 with NGINX
1. SSH into your EC2 instance
ssh -i your-key.pem ubuntu@your-ec2-ip
2. Install Docker & NGINX
sudo apt update
sudo apt upgrade -y
curl -o get-docker.sh https://get.docker.com/
sudo bash get-docker.sh
sudo apt install nginx -y
3. Copy the project or push to EC2
You can SCP or clone the repo directly into EC2.
git clone https://github.com/vasiliskou/multimodal-assistant.git
4. Configure NGINX
Create /etc/nginx/sites-available/gradio:
server {
listen 80;
server_name your_domain;
location / {
proxy_pass http://localhost:7860;
proxy_set_header Host $host;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection 'upgrade';
proxy_cache_bypass $http_upgrade;
}
}
Enable the config:
sudo ln -s /etc/nginx/sites-available/gradio /etc/nginx/sites-enabled/
sudo nginx -t
sudo systemctl restart nginx
5. (Optional) Secure with HTTPS
If you don't have a domain yet, you can register one using AWS Route 53 and point it to your EC2 instance via an A record.
Once your domain is set up and pointing to your EC2 public IP:
Install Certbot and the NGINX plugin:
sudo apt install certbot python3-certbot-nginx -y
Obtain and install a free SSL certificate from Let's Encrypt:
sudo certbot --nginx -d your_domain
Replace
your_domainwith your actual domain.
📂 Project Structure
.
├── app.py # Main Gradio app
├── config.yaml # Model + API configuration
├── Dockerfile # Docker build instructions
├── requirements.txt # Python dependencies
├── run.sh # Shell script to run the app
├── README.md # Project documentation
├── .gitignore # Git ignored files
├── .dockerignore # Docker ignored files