Feynmagi is a multi-modal LLM interactive agents solution.
Project description
Feynmagi V 0.1.31
Install ollama
- curl -fsSL https://ollama.com/install.sh | sh
- or go to : https://ollama.com/download
Install feynmagi
- pip install feynmagi
Or install from repo
- git clone https://github.com/Imag2020/feynmagi
- cd feynmagi
- pip install .
How to run
- feynmagi
- You can access it from any browser at localhost:5000
🌟 Feynmagi is a multi-modal tool for MLLM interactive agents that integrates all the latest AI technologies into a single application, capable of running models locally on CPU or GPU (via Ollama). It incorporates proprietary code for embedding, RAG, and interactive AI, designed to model intelligent AIs with the goal of experimenting with AGI.
In this alpha test version 0.1.30, we add:
- Agents management with next agent call
- UI messages I/O display update with atatch button
In this alpha test version 0.1.30, we add:
- Agents management with live run and cron scheduler
- swicth to gemma2 default local llm
- test tolls and fucntion calling with groq
- use of florence-2 for images
- send mail via smtp server (todo ==> use gmail OAuth)
- NB : mail config json under data dir, rename config_mail_to_do.json to config_mail.json
In this alpha test version 0.1.26, we have:
- Integration with OpenAI and Groq clients for external and faster inference, alongside local inference using Ollama.
- Chat capabilities with LLMs, images, documents, or AI agents.
- Ingestion of documents or web pages into a local vector database.
- Retrieval of past session histories.
- Support for new models in Ollama without code modification.
- use of florence-2 as vision model
The following agents are under development:
- Web search agent via Google 🌐
- Code interpreter agent ⚙️
- Memory agent 🧠
⚠️ Please note that this code is experimental and in the testing phase. It should not be tested on enterprise or confidential data. This code is shared for the AI community within an academic framework for scientific research, and the author does not authorize its use outside this scope without prior permission.
We welcome any contributions to the development towards this stated goal.
Visit us at feynmagi.com or email us at imed@feynmagi.com.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file feynmagi-0.1.31.tar.gz.
File metadata
- Download URL: feynmagi-0.1.31.tar.gz
- Upload date:
- Size: 12.1 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.10.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
db6b4211ee58821f067b68fa76d6c5379b1a4cbd46c2963988006eeb0047e362
|
|
| MD5 |
b04ceeb725d433a91a1aa72beb482fe6
|
|
| BLAKE2b-256 |
e49e1aa61b6954e5841b5e484d6619d560de32a8fc866963e6745dd20f927b7f
|
File details
Details for the file feynmagi-0.1.31-py3-none-any.whl.
File metadata
- Download URL: feynmagi-0.1.31-py3-none-any.whl
- Upload date:
- Size: 13.0 MB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.10.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
55c51fa7d6afca760738a3e09765053b99c4a2067d40c67a4296ec996cd97541
|
|
| MD5 |
f41be3a0d3a50194ba57998e752b0d2a
|
|
| BLAKE2b-256 |
aa42c140c377cb80fc27e31f8a0301828d3c54179bd40418404d60ade488491e
|