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, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
Apache License 2.0
12.05k
stars
827
forks
source link
Added redirection to wiki, for the repo - as this repo still maintain… #186
The main repos maintains the top spot for SEO, having a redirection to a more "new user friendly" site will help with onboarding. Instead of overwhelming them
The main repos maintains the top spot for SEO, having a redirection to a more "new user friendly" site will help with onboarding. Instead of overwhelming them