LiteLLM
LiteLLM gives you a unified interface to access multiple LLMs, track LLM usage, setup guardrails, custom logging and more.
Installation
First, make sure you have the litellm library installed, if not you can install it by running:
pip install litellmBasic Completion Example
This example shows you how to make a simple completion request to a Featherless AI model.
Set the FEATHERLESS_API_KEY environment variable to your Featherless API key before running either example. In the model value, featherless_ai/ is LiteLLM's provider-routing prefix; deepseek-ai/DeepSeek-V4.1-Flash-1017 is the exact model ID from the Featherless catalogue. Keep both parts when calling LiteLLM.
import os
from litellm import completion
response = completion(
model="featherless_ai/deepseek-ai/DeepSeek-V4.1-Flash-1017",
api_key=os.environ["FEATHERLESS_API_KEY"],
messages=[{"role": "user", "content": "Write a short poem about AI."}]
)
print(response.choices[0].message.content)Streaming Example
LiteLLM also supports streaming responses
import os
from litellm import completion
response_stream = completion(
model="featherless_ai/deepseek-ai/DeepSeek-V4.1-Flash-1017",
api_key=os.environ["FEATHERLESS_API_KEY"],
messages=[{"role": "user", "content": "Tell me a fun fact about space"}],
stream=True
)
print("Streaming response:")
for chunk in response_stream:
if chunk.choices and chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="", flush=True)
print("\n\nStreaming complete.")Resources
LiteLLM Featherless docs
The LiteLLM Featherless documentation
LiteLLM Featherless Cookbook
A notebook to guide you on how to setup LiteLLM with Featherless
Model Catalog
Our catalog of models