Tools

AI Gallery

Tools

AI Gallery

The full-stack AI infra

Serve models, train on GPUs, Host apps

The full-stack AI infrastructure

Serve models, train on GPUs, Host apps

Trusted by Engineers

Trusted by Engineers

Why Use /ML

Why /ML?

Don't waste time fighting configs
We have a Docker and a UI for EVERYTHING

Serve Large Language models

Serve Large Language models

Serve Large Language models

All you have to do is make sure its available on hugging-face

All you have to do is make sure its available on hugging-face

# Serve Large Language Models
def serve_llm(model_name, input_text):
    # Initialize the model
    model = load_model(model_name)
    # Generate response
    response = model.generate(input_text)
    # Log for observability
    log_inference(model_name, input_text, response)
    return response
# Serve Large Language Models
def serve_llm(model_name, input_text):
    # Initialize the model
    model = load_model(model_name)
    # Generate response
    response = model.generate(input_text)
    # Log for observability
    log_inference(model_name, input_text, response)
    return response

Serve and train Multi-modal models

Train Multi-modal models

on /ML Workspaces or Your workspace

in other words, our GPUs or Yours

on /ML Workspaces or Your workspace

in other words, our GPUs or Yours

def train_multimodal(model_name, dataset, epochs=10):
    model = load_multimodal_model(model_name)
    model.train(dataset, epochs=epochs)
    log_training_metrics(model_name, model.metrics)
    model.save()
    return model
def train_multimodal(model_name, dataset, epochs=10):
    model = load_multimodal_model(model_name)
    model.train(dataset, epochs=epochs)
    log_training_metrics(model_name, model.metrics)
    model.save()
    return model

Host your Streamlit, Gradio and Dash Apps

Host your Streamlit, Gradio and Dash Apps

Share your dashboards and apps with users

Share your dashboards and apps with users

# Host Streamlit, Gradio and Dash Apps
def deploy_app(app_type, app_path):
    if app_type == "streamlit":
        cmd = f"streamlit run {app_path} --server.enableCORS=false"
    elif app_type == "gradio":
        cmd = f"gradio {app_path}"
    elif app_type == "dash":
        cmd = f"python {app_path}"
    
    # Deploy with observability
    process = start_process(cmd)
    setup_monitoring(process.pid, app_type)
    return f"App deployed at http://your_domain:{process.port}"
# Host Streamlit, Gradio and Dash Apps
def deploy_app(app_type, app_path):
    if app_type == "streamlit":
        cmd = f"streamlit run {app_path} --server.enableCORS=false"
    elif app_type == "gradio":
        cmd = f"gradio {app_path}"
    elif app_type == "dash":
        cmd = f"python {app_path}"
    
    # Deploy with observability
    process = start_process(cmd)
    setup_monitoring(process.pid, app_type)
    return f"App deployed at http://your_domain:{process.port}"


Cost observability

Increase cloud cost visibility and gain clear insights into cloud spending

Ready to scale your AI?

Ready to scale you AI?

©2024 – Made with ❤️ & ☕️ in Montreal

©2024 – Made with ❤️ & ☕️ in Montreal

©2024 – Made with ❤️ & ☕️ in Montreal