gitpulse

BlinkDL/RWKV-LM

RWKV (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.

14,778 星标1,027 复刻PythonApache-2.0推送于 2026-11-04

在 GitHub 上查看 →

星标动能

今日+0 (+0.0%)
本周+6 (+0.0%)
本月+30 (+0.2%)

主题

attention-mechanismdeep-learninggptgpt-2gpt-3language-modellinear-attentionlstmpytorchrnntransformertransformersrwkvchatgpt