Skip to main content

Documentation Index

Fetch the complete documentation index at: https://sambanova-systems.mintlify.dev/docs/llms.txt

Use this file to discover all available pages before exploring further.

Gradio is an open-source python library for building interactive web applications to showcase machine learning models, APIs, and python functions. The sambanova-gradio package extends Gradio’s capabilities by simplifying SambaNova model integration. It provides a user-friendly interface for developers to connect SambaNova models to their applications. The sambanova-gradio package has two dependencies: OpenAI and Gradio. It streamlines the setup process with a built-in registry function, sambanova_gradio.registry, which facilitates seamless connections to various models. The registry function accepts a model ID as the name parameter. Executing the script automatically generates an application interface using the specified model.; For implementation details and additional configurations, see the GitHub repository.

​
Prerequisites

  • A SambaCloud account
  • An API key from your SambaNova dashboard

​
Installation

Follow these steps to integrate SambaNova models with Gradio.
1

Install the sambanova-gradio package by running the command below to install the required package.

pip install sambanova-gradio
2

Set Your SambaNova API Key by saving your API key as an environment variable, as shown below.

export SAMBANOVA_API_KEY=<your token>

​
Create a Gradio interface

Use the sambanova_gradio.registry function to build and launch your Gradio application, as shown below.
Running this Python script will launch a Gradio interface connected to the Meta-Llama-3.3-70B-Instruct model.
import gradio as gr
import sambanova_gradio

gr.load(
    name='Meta-Llama-3.3-70B-Instruct',
    src=sambanova_gradio.registry,
).launch()

​
Customize your interface

You can personalize the interface by modifying the model, input/output components, and additional settings to gr.Interface, as shown below. 
import gradio as gr
import sambanova_gradio

gr.load(
    name='Meta-Llama-3.3-70B-Instruct',
    src=sambanova_gradio.registry,
    title='Sambanova-Gradio Integration',
    description="Chat with Meta-Llama-3.3-70B-Instruct model.",
    examples=["Explain quantum gravity to a 5-year old.", "How many R are there in the word Strawberry?"]
).launch()

​
Example screenshot

The following interface was generated using the above code snippet. Gradio chat interface with Meta-Llama-3.3-70B-Instruct See more examples of Gradio applications in the SambaNova HuggingFace space.