#devices #edge #tensor #ai #tpu #future #framework

phynexus

Empowering Intelligent Futures, One Edge at a Time

2 stable releases

new 1.0.1 Mar 15, 2025

#314 in Machine learning

Apache-2.0

130KB
3K SLoC

Neurenix

Neurenix is ​​an AI framework optimized for embedded devices (Edge AI), with support for multiple GPUs and distributed clusters. The framework specializes in AI agents, with native support for multi-agent, reinforcement learning, and autonomous AI.

Social

Bluesky Discord GitHub Mastodon X/Twitter

Main Features

  • Phynexus Engine: High-performance engine for tensor operations
  • Optimized for Edge AI: Efficient execution on embedded devices
  • Multi-GPU and Distributed Support: Scalability to clusters and multiple GPUs
  • TPU Support: Tensor Processing Unit acceleration for machine learning workloads
  • Agent-Specialized: Native support for multi-agent and autonomous AI
  • Declarative API: Simpler and more intuitive interface than traditional frameworks
  • LLM Fine-tuning: Simplified tools for tuning language models
  • Hugging Face Integration: Compatibility with pre-existing models

Documentation

Neurenix Documentation

License

This project is licensed under the Apache License 2.0.


lib.rs:

Phynexus: A high-performance tensor library for embedded devices

Dependencies

~11–20MB
~307K SLoC