Developers around the world have been using SambaCloud to create amazing applications using the best open-source models running with fast and efficient inference, thanks to the SambaNova RDU. SambaCloud offers an OpenAI-compatible REST API that makes it extremely easy for developers to start building within minutes.
The SambaNova SDK
Today, we are unveiling the SambaNova SDK to make it even simpler to get started building — providing native, type-safe SDKs for both Python and JavaScript/TypeScript making it easy to interact with the SambaNova REST API. You will have access to a range of models on SambaCloud including DeepSeek, Llama and Qwen. Each has its own capabilities, such as text image or audio processing, to support your AI applications.
Here are some of the popular models used by developers today:
- DeepSeek-V3.1-Terminus
- gpt-oss-120b
- Llama-4-Maverick-17B-128E-Instruct
- Whisper-Large-v3
- Qwen3-32B
Check this link out for the full list.
Building in seconds…
Step 1: Installation
To get started in Python, you can install the SDK from PyPI with a single command.
Open a terminal window and install the SambaNova SDK.
# make sure you have Python3 and pip installed
pip install sambanova
Step 2: Setup and authentication
Before you can call any models, you’ll need:
- Your API key from the SambaNova platform
- The base URL for your deployment (e.g., https://api.sambanova.ai/v1)
Then initialize the client:
from sambanova import SambaNova
client = SambaNova(
api_key="your-sambanova-api-key",
base_url="https://api.sambanova.ai/v1",
)
That’s it — you’re now ready to interact with any of the supported models.
Step 3: Your first chat completion
completion = client.chat.completions.create(
model="DeepSeek-V3.1-Terminus",
messages=[{"role":"system","content":"You are a helpful assistant"},{"role":"user","content":"Share a happy story with me"}],
temperature=0.1,
top_p=0.1
)
print(completion.choices[0].message.content)
The SambaNova API and SDK support the wide range of tuning parameters offered by every model. Here are some of the parameters used above and how you can best control them.
temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
top_p: An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.
top_k: The top_k parameter is used to limit the number of choices for the next predicted word or token.
SDK functionality extended
Beyond just chat completions, the SDK supports many other methods to make it easier to work with different types of models and modalities. Below is the full list and you can review the linked examples:
- Chat Completions: sambanova.chat.completions.create()
- Creates chat-based completions
- Embeddings: sambanova.embeddings.create()
- Creates vector embeddings
- Transcriptions: sambanova.audio.transcriptions.create()
- Transcribes audio into the input language
- Translations: sambanova.audio.translations.create()
- Translated audio into English
- The Full SambaCloud Model List: sambanova.models.list()
- Gets the available model list metadata
- Model Metadata: sambanova.models.retrieve(model_id)
- Gets metadata for a specific model
The SambaNova Python SDK gives developers a clear, intuitive way to build intelligent systems with top-tier models. If you can import a package and write a few lines of Python, you’re ready to build production-ready AI apps.
Check out this Google Colab Notebook that is publicly available to get started here.
Final thoughts
The SambaNova SDK is more than a wrapper, it’s a bridge between cutting-edge AI models and the tools you already use. Whether you’re building a chat app, a retrieval pipeline, or a multimodal agent, the SDK gives you a clean, powerful interface to bring those ideas to life.
Where to go next
Explore the SambaNova SDK docs for full API details. Code examples are also included for:
- Text generation: Use various methods to generate text, including non-streaming, streaming, and asynchronous completions.
- Function calling: Enables dynamic workflows by allowing the model to select and suggest function calls based on user input, which helps in building agentic workflows.
- Audio: For developers requiring audio support, SambaNova provides OpenAI’s Whisper large-v3 model, which enables real-time transcriptions and translations.
- Vision: Vision models support multimodal inputs, allowing users to process both text and images. These models analyze images and generate context-aware text responses.
- Embeddings: An embedding is a vectorized numerical representation of data, where the relatedness between different pieces of information is captured through measurable distances. These embeddings are commonly used in applications such as recommendation systems, classification, and search.
Github Repository for the SambaNova Python SDK.
Github Repository for the SambaNova TypeScript SDK.
Sign up for the SambaCloud here.
About SambaCloud
SambaCloud is a powerful platform that enables developers to easily integrate the best open-source models with the fastest inference speeds. Powered by our state-of-the-art AI chip, the SN40L, SambaCloud provides a seamless and efficient way to build AI applications with fast inference. Get started today and experience the benefits of fast inference speeds, maximum accuracy, and an enhanced developer experience, in just three easy steps!