RWKV-LM
PublicRWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
Creat：2021-08-08T14:05:27
Update：2025-03-26T16:45:01
13.5K
Stars
3
Stars Increase