Model talk: 1.5G memory can run RNN 14B open source model (ChatRWKV)

Original link: https://soulteary.com/2023/03/25/model-talk-open-source-model-of-rnn-14b-that-can-run-on-little-gpu-memory-chatrwkv.html

docker-test.jpg

In this article, let’s talk about how to quickly get started with a model with 14B parameters, but a special RNN model: ChatRWKV. This article will introduce how to get started quickly, including using a 4090 high-speed inference generation content with 24 video memory, and how to run this model with only 1.5G video memory.

This article is transferred from: https://soulteary.com/2023/03/25/model-talk-open-source-model-of-rnn-14b-that-can-run-on-little-gpu-memory-chatrwkv.html
This site is only for collection, and the copyright belongs to the original author.