Setup

To be able monitor the runs of your agent, we are first hooking into your running system. For this, you have to provide us with your model providers API key.

Import Your Model Provider

In the sidebar, go to Providers and click Import Provider.

Then select your model provider and enter your API key

Connect Your Agent To Augento

After you imported your model provider, we now can connect your agent to Augento.

In the sidebar, go to Models and click New Model. Give your new model a name and select the provider we just created.

After that, choose a name for the first API key for the newly created model.

Augento API keys are tied to a specific model and can only be used with that model.

Copy the API key beginning with sk-au-.

Now in the code of your agent, find the place where you instantiate your LLM connector, replace the api-key and change the base_url to https://api.augento.ai/v1:

Python - Langchain

agent.py


from langchain_openai import ChatOpenAI
 
llm = ChatOpenAI(
    api_key="sk-au-...",
    base_url="https://api.augento.ai/v1"
)

Python - OpenAI

agent.py


from openai import OpenAI
 
llm = OpenAI(
    api_key="sk-au-...",
    base_url="https://api.augento.ai/v1"
)

JS/TS - OpenAI

agent.ts


import { OpenAI } from "openai";
 
const openai = new OpenAI({
  apiKey: "sk-au-...",
  baseURL: "https://api.augento.ai/v1"
});

Model Selection

You can keep using all your model provider’s models until you have trained your first model using Augento. For example, if you are using OpenAI, you can use the gpt-4o model:

Python - Langchain

agent.py


from langchain_openai import ChatOpenAI
 
llm = ChatOpenAI(
    model="gpt-4o",
    api_key="sk-au-...",
    base_url="https://api.augento.ai/v1"
)

Python - OpenAI

agent.py


from openai import OpenAI
 
llm = OpenAI(
    api_key="sk-au-...",
    base_url="https://api.augento.ai/v1"
)
 
completion = llm.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)

JS/TS - OpenAI

agent.ts


import { OpenAI } from "openai";
 
const openai = new OpenAI({
  apiKey: "sk-au-...",
  baseURL: "https://api.augento.ai/v1"
});
 
completion = openai.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
);

Augento uses a special syntax to select models that have been fine-tuned on our platform. Learn more about model selection here.