From 1f189a4034c7ff1d0c145ef556327de1fa90da51 Mon Sep 17 00:00:00 2001
From: PENG Bo <33809201+BlinkDL@users.noreply.github.com>
Date: Tue, 22 Mar 2022 04:55:04 +0800
Subject: [PATCH] Update README.md

---
 README.md | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/README.md b/README.md
index cd0b010..765f78f 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,21 @@
 # The RWKV Language Model
 
+## v2
+
+RWKV v2 is a RNN which can also be directly trained like a GPT transformer.
+
+You only need x_t, a_t, b_t of position t to compute the vectors for position t+1. 
+
+Hence it can be 100x faster than GPT, and 100x more VRAM friendly.
+
+I AM STILL TRAINING A LM TO TEST ITS CONVERGENCE.
+
+The model:
+
+![RWKV-v2-RNN](RWKV-v2-RNN.png)
+
+## v1
+
 We propose the RWKV language model, with alternating time-mix and channel-mix layers:
 
 <img src=