Metadata-Version: 2.4
Name: ai-microcore
Version: 4.0.0.dev21
Summary: # Minimalistic Foundation for AI Applications
Keywords: llm,large language models,ai,similarity search,ai search,gpt,openai,framework,adapter
Author-email: Vitalii Stepanenko <mail@vitalii.in>
Maintainer-email: Vitalii Stepanenko <mail@vitalii.in>
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Developers
Classifier: Topic :: Software Development :: Libraries :: Application Frameworks
License-File: LICENSE
Requires-Dist: openai>=0.28.1,<2
Requires-Dist: python-dotenv~=1.1.0
Requires-Dist: Jinja2~=3.1.2
Requires-Dist: colorama~=0.4.6
Requires-Dist: PyYAML~=6.0
Requires-Dist: chardet~=5.2.0
Requires-Dist: tiktoken>=0.7.0,<1.0
Requires-Dist: mcp~=1.9.2
Requires-Dist: fastmcp~=2.8.0
Requires-Dist: docstring_parser~=0.16.0
Project-URL: Source Code, https://github.com/Nayjest/ai-microcore

<p align="right"> 
    <a href="https://github.com/Nayjest/ai-microcore/releases" target="_blank"><img src="https://img.shields.io/github/v/release/Nayjest/ai-microcore.svg" alt="Release Notes"></a>
    <a href="https://app.codacy.com/gh/Nayjest/ai-microcore/dashboard?utm_source=gh&utm_medium=referral&utm_content=&utm_campaign=Badge_grade" target="_blank"><img src="https://app.codacy.com/project/badge/Grade/441d03416bc048828c649129530dcbc3" alt="Code Quality"></a>
    <a href="https://github.com/Nayjest/ai-microcore/actions/workflows/pylint.yml" target="_blank"><img src="https://github.com/Nayjest/ai-microcore/actions/workflows/pylint.yml/badge.svg" alt="Pylint"></a>
    <img src="coverage.svg" alt="Code Coverage">
    <a href="https://github.com/Nayjest/ai-microcore/actions/workflows/tests.yml" target="_blank"><img src="https://github.com/Nayjest/ai-microcore/actions/workflows/tests.yml/badge.svg" alt="Tests"></a>
    <a href="https://github.com/Nayjest/ai-microcore/blob/main/LICENSE" target="_blank"><img src="https://img.shields.io/static/v1?label=license&message=MIT&color=d08aff" alt="License"></a>
</p>


# AI MicroCore: A Minimalistic Foundation for AI Applications

**MicroCore** is a collection of python adapters for Large Language Models
and Vector Databases / Semantic Search APIs allowing to 
communicate with these services in a convenient way, make them easily switchable 
and separate business logic from the implementation details.

It defines interfaces for features typically used in AI applications,
which allows you to keep your application as simple as possible and try various models & services
without need to change your application code.

You even can switch between text completion and chat completion models only using configuration.

Thanks to LLM-agnostic MCP integration,
**MicroCore** connects MCP tools to any language models easily,
whether through API providers that do not support MCP, or through inference using pytorch or arbitrary python functions.

The basic example of usage is as follows:

```python
from microcore import llm

while user_msg := input('Enter message: '):
    print('AI: ' + llm(user_msg))
```

## 🔗 Links

 -   [API Reference](https://ai-microcore.github.io/api-reference/)
 -   [PyPi Package](https://pypi.org/project/ai-microcore/)
 -   [GitHub Repository](https://github.com/Nayjest/ai-microcore)


## 💻 Installation

Install as PyPi package:
```
pip install ai-microcore
```

Alternatively, you may just copy `microcore` folder to your project sources root.
```bash
git clone git@github.com:Nayjest/ai-microcore.git && mv ai-microcore/microcore ./ && rm -rf ai-microcore
```


## 📋 Requirements

Python 3.10 / 3.11 / 3.12 / 3.13

Both v0.28+ and v1.X OpenAI package versions are supported.


## ⚙️ Configuring

### Minimal Configuration

Having `OPENAI_API_KEY` in OS environment variables is enough for basic usage.

Similarity search features will work out of the box if you have the `chromadb` pip package installed.

### Configuration Methods

There are a few options available for configuring microcore:

-   Use `microcore.configure(**params)`
    <br>💡 <small>All configuration options should be available in IDE autocompletion tooltips</small>
-   Create a `.env` file in your project root; examples: [basic.env](https://github.com/Nayjest/ai-microcore/blob/main/.env.example), [Mistral Large.env](https://github.com/Nayjest/ai-microcore/blob/main/.env.mistral.example), [Anthropic Claude 3 Opus.env](https://github.com/Nayjest/ai-microcore/blob/main/.env.anthropic.example), [Gemini on Vertex AI.env](https://github.com/Nayjest/ai-microcore/blob/main/.env.google-vertex-gemini.example), [Gemini on AI Studio.env](https://github.com/Nayjest/ai-microcore/blob/main/.env.gemini.example)
-   Use a custom configuration file: `mc.configure(DOT_ENV_FILE='dev-config.ini')`
-   Define OS environment variables

For the full list of available configuration options, you may also check [`microcore/config.py`](https://github.com/Nayjest/ai-microcore/blob/main/microcore/config.py).

### Installing vendor-specific packages
For the models working not via OpenAI API, you may need to install additional packages:
#### Anthropic Claude 3
```bash
pip install anthropic
```
#### Google Gemini via AI Studio
```bash
pip install google-generativeai
```
#### Google Gemini via Vertex AI
```bash
pip install vertexai
```
📌Additonaly for working through [Vertex AI](https://cloud.google.com/vertex-ai) you need to
[install the Google Cloud CLI](https://cloud.google.com/sdk/docs/install)
and [configure the authorization](https://cloud.google.com/sdk/docs/authorizing).

#### Local language models via Hugging Face Transformers

You will need to install transformers and deep learning library of your choice (PyTorch, TensorFlow, Flax, etc).

See [transformers installation](https://huggingface.co/docs/transformers/installation).

### Priority of Configuration Sources

1.  Configuration options passed as arguments to `microcore.configure()` have the highest priority.
2.  The priority of configuration file options (`.env` by default or the value of `DOT_ENV_FILE`) is higher than OS environment variables.
    <br>💡 <small>Setting `USE_DOT_ENV` to `false` disables reading configuration files.</small>
3.  OS environment variables have the lowest priority.

### Vector Databases

Vector database functions are available via `microcore.texts`.
Default vector database is [Chroma](https://www.trychroma.com/).
In order to use vector database functions, you need to install the `chromadb` package:
```bash
pip install chromadb
```
By default, MicroCore will use ChromaDB PersistentClient (if corresponding package is installed).
Alternatively, you can run Chroma as separate service and configure MicroCore to use HttpClient:

```python
from microcore import configure
configure(
    EMBEDDING_DB_HOST = 'localhost',
    EMBEDDING_DB_PORT = 8000,
)
```

## 🌟 Core Functions

### llm(prompt: str, \*\*kwargs) → str

Performs a request to a large language model (LLM).

Asynchronous variant: `allm(prompt: str, **kwargs)`

```python
from microcore import *

# Will print all requests and responses to console
use_logging()

# Basic usage
ai_response = llm('What is your model name?')

# You also may pass a list of strings as prompt
# - For chat completion models elements are treated as separate messages
# - For completion LLMs elements are treated as text lines
llm(['1+2', '='])
llm('1+2=', model='gpt-4')

# To specify a message role, you can use dictionary or classes
llm(dict(role='system', content='1+2='))
# equivalent
llm(SysMsg('1+2='))

# The returned value is a string
assert '7' == llm([
 SysMsg('You are a calculator'),
 UserMsg('1+2='),
 AssistantMsg('3'),
 UserMsg('3+4=')]
).strip()

# But it contains all fields of the LLM response in additional attributes
for i in llm('1+2=?', n=3, temperature=2).choices:
    print('RESPONSE:', i.message.content)

# To use response streaming you may specify the callback function:
llm('Hi there', callback=lambda x: print(x, end=''))

# Or multiple callbacks:
output = []
llm('Hi there', callbacks=[
    lambda x: print(x, end=''),
    lambda x: output.append(x),
])
```

### tpl(file_path, \*\*params) → str
Renders prompt template with params.

Full-featured Jinja2 templates are used by default.

Related configuration options:

```python
from microcore import configure
configure(
    # 'tpl' folder in current working directory by default
    PROMPT_TEMPLATES_PATH = 'my_templates_folder'
)
```

### texts.search(collection: str, query: str | list, n_results: int = 5, where: dict = None, **kwargs) → list[str]
Similarity search

### texts.find_one(self, collection: str, query: str | list) → str | None
Find most similar text

### texts.get_all(self, collection: str) -> list[str]
Return collection of texts

### texts.save(collection: str, text: str, metadata: dict = None))
Store text and related metadata in embeddings database

### texts.save_many(collection: str, items: list[tuple[str, dict] | str])
Store multiple texts and related metadata in the embeddings database

### texts.clear(collection: str):
Clear collection

## API providers and models support

LLM Microcore supports all models & API providers having OpenAI API.

### List of API providers and models tested with LLM Microcore:

| API Provider                                                                             |                                                                                                                                      Models |
|------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------:|
| [OpenAI](https://openai.com)                                                             |                                    All GPT-4 and GTP-3.5-Turbo models<br/>all text completion models (davinci, gpt-3.5-turbo-instruct, etc) |
| [Microsoft Azure](https://azure.microsoft.com/en-us/products/ai-services/openai-service) |                                                                                                            All OpenAI models, Mistral Large |
| [Anthropic](https://anthropic.com)                                                       |                                                                                                                             Claude 3 models |
| [MistralAI](https://mistral.ai)                                                          |                                                                                                                          All Mistral models |
| [Google AI Studio](https://aistudio.google.com/)                             |                                                                                                                        Google Gemini models |
| [Google Vertex AI](https://cloud.google.com/vertex-ai?hl=en)                             |                                                   Gemini Pro & [other models](https://cloud.google.com/vertex-ai/docs/start/explore-models) |
| [Deep Infra](https://deepinfra.com)                                                      | deepinfra/airoboros-70b<br/>jondurbin/airoboros-l2-70b-gpt4-1.4.1<br/>meta-llama/Llama-2-70b-chat-hf<br/>and other models having OpenAI API |
| [Anyscale](https://anyscale.com)                                                         |                                           meta-llama/Llama-2-70b-chat-hf<br/>meta-llama/Llama-2-13b-chat-hf<br/>meta-llama/Llama-7b-chat-hf |
| [Groq](https://groq.com/)                                                         |                                           LLaMA2 70b<br>Mixtral 8x7b<br>Gemma 7b |
| [Fireworks](fireworks.ai)                                                         |                                           [Over 50 open-source language models](https://fireworks.ai/models?show=All) |

## Supported local language model APIs:
- HuggingFace [Transformers](https://huggingface.co/docs/transformers/index) (see configuration examples [here](https://github.com/Nayjest/ai-microcore/blob/main/tests/local/test_transformers.py)).
- Custom local models by providing own function for chat / text completion, sync / async inference.

## 🖼️ Examples

#### [Code review tool](https://github.com/llm-microcore/microcore/blob/main/examples/code-review-tool)
Performs code review by LLM for changes in git .patch files in any programming languages.

#### [Image analysis](https://colab.research.google.com/drive/1qTJ51wxCv3VlyqLt3M8OZ7183YXPFpic) (Google Colab)
Determine the number of petals and the color of the flower from a photo (gpt-4-turbo)

#### [Banchmark LLMs on math problems](https://www.kaggle.com/code/nayjest/gigabenchmark-llm-accuracy-math-problems) (Kaggle Notebook)
Benchmark accuracy of 20+ state of the art models on solving olympiad math problems. Inferencing local language models via HuggingFace Transformers, parallel inference.
 
#### [Other examples](https://github.com/llm-microcore/microcore/tree/main/examples)

## Python functions as AI tools

@TODO

## 🤖 AI Modules
**This is experimental feature.**

Tweaks the Python import system to provide automatic setup of MicroCore environment
based on metadata in module docstrings.
### Usage:
```python
import microcore.ai_modules
```
### Features:

*   Automatically registers template folders of AI modules in Jinja2 environment

## 🛠️ Contributing

Please see [CONTRIBUTING](https://github.com/Nayjest/ai-microcore/blob/main/CONTRIBUTING.md) for details.


## 📝 License

Licensed under the [MIT License](https://github.com/Nayjest/ai-microcore/blob/main/LICENSE)
© 2023 [Vitalii Stepanenko](mailto:mail@vitaliy.in)

