monoai

The complete framework to build AI-powered applications

MonoAI is a Python library that enables structured, standardized, and efficient interactions with multiple AI models, harnessing their collective intelligence to deliver richer, more accurate responses.

🚀 Quick Start

Prompts

In MonoAI, a prompt is the request you send to a model.
You can define prompts in three different ways:

1. Plain Text

Directly write the prompt inside the ask method:

from MonoAI.models import Model

model = Model()
response = model.ask("What is 2 + 2?")

2. Prompt Class

Use the Prompt class to enhance your prompts with additional parameters such as response types, chaining, and iteration control:

from MonoAI.models import Model
from MonoAI.prompts import Prompt

model = Model()
prompt = Prompt("What is 2 + 2?", response_type=int)
response = model.ask(prompt)

3. .prompt Files

Store prompts in external .prompt files to separate your codebase from content logic.

Example - sum.prompt:

What is 2 + 2?

from MonoAI.models import Model
from MonoAI.prompts import Prompt

model = Model()
prompt = Prompt(prompt_id="sum", response_type=int)
response = model.ask(prompt)

You can also define additional metadata using an XML-like syntax inside .prompt files:

Example with metadata:

# sum.prompt
<prompt response_type="int">
What is 2 + 2?
</prompt>

Models

MonoAI provides multiple model interfaces to handle different use cases, from basic generation to multi-model aggregation:

Model

The standard interface for interacting with a single AI model.

MultiModel

Send a prompt to multiple models asynchronously and retrieve their individual outputs:

from MonoAI.models import MultiModel

model = MultiModel(models=[
    {"provider": "openai", "model": "gpt-4o"},
    {"provider": "deepseek", "model": "chat"}
])
response = model.ask("What is the capital of Italy?")

CollectiveModel

Send a prompt to multiple models and aggregate their outputs using a separate aggregator model for a richer, consolidated answer:

from MonoAI.models import CollectiveModel

model = CollectiveModel(
    models=[
        {"provider": "openai", "model": "gpt-4o"},
        {"provider": "deepseek", "model": "chat"}
    ],
    aggregator={"provider": "openai", "model": "gpt-4o"}
)
response = model.ask("What is the capital of Italy?")

AutoModel

Automatically select the best model for a given prompt based on configuration or inference:

from MonoAI.models import AutoModel

model = AutoModel()
response = model.ask("What is the capital of Italy?")

🔑 API Key Management

MonoAI simplifies API key management through a providers.keys file in the root directory.
Each line should follow this format:

PROVIDER_NAME=API_KEY

Example providers.keys:

OPENAI=sk-proj-ABCDE12345
DEEPSEEK=sk-proj-FGHIJ67890

MonoAI automatically loads these keys at runtime — no extra setup needed.

⚙️ Configuration

Configure MonoAI globally via a monoai.yaml file.
Supported fields:

Example monoai.yaml:

prompts_path: prompts
keysfile_path: providers.keys
base_model:
  provider: openai
  model: gpt-4o-mini