RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference,…

ChatRWKV is like ChatGPT but powered by RWKV (100% RNN) language model, and open source.

BELLE: Be Everyone's Large Language model Engine（开源中文对话大模型）

Paper, Code and Statistics for Self-Supervised Learning and Pre-Training on Speech.

Moved to https://github.com/k2-fsa/icefall

FSA/FST algorithms, differentiable, with PyTorch compatibility.