LangChain + Aethon

Aethon is OpenAI-compatible, so LangChain's ChatOpenAI works out of the box — just point it at Aethon's base URL.

Install

bash

pip install langchain-openai

Basic Usage

python

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    base_url="https://api.aethon.ai/v1",
    api_key="your-aethon-key",
    model="llama3.1:8b",
)

response = llm.invoke("Explain GPU compute in one sentence")
print(response.content)

Streaming

python

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    base_url="https://api.aethon.ai/v1",
    api_key="your-aethon-key",
    model="llama3.1:8b",
    streaming=True,
)

for chunk in llm.stream("Write a haiku about GPUs"):
    print(chunk.content, end="", flush=True)

With Chains

python

from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate

llm = ChatOpenAI(
    base_url="https://api.aethon.ai/v1",
    api_key="your-aethon-key",
    model="llama3.1:8b",
)

prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant that explains things simply."),
    ("user", "{question}"),
])

chain = prompt | llm

response = chain.invoke({"question": "What is a GPU?"})
print(response.content)

With Embeddings

python

from langchain_openai import OpenAIEmbeddings

embeddings = OpenAIEmbeddings(
    base_url="https://api.aethon.ai/v1",
    api_key="your-aethon-key",
    model="nomic-embed-text",
)

vectors = embeddings.embed_documents(["Hello world", "GPU computing"])
print(f"Dimension: {len(vectors[0])}")

No wrapper package needed. Since Aethon is OpenAI-compatible, you use LangChain's built-inlangchain-openaipackage. Just change the base_url and api_key.

LlamaIndex Guide →LiteLLM Guide →Open WebUI Guide →