https://github.com/arrmansa/basic-ui-for-gpt-neo-with-low-vram

A basic ui for running gpt neo 2.7B on low vram (3 gb Vram minimum)
https://github.com/arrmansa/basic-ui-for-gpt-neo-with-low-vram

gpt gpt-3 gpt-neo transformers

Last synced: 6 months ago
JSON representation

A basic ui for running gpt neo 2.7B on low vram (3 gb Vram minimum)

• Host: GitHub
• URL: https://github.com/arrmansa/basic-ui-for-gpt-neo-with-low-vram
• Owner: arrmansa
• License: apache-2.0
• Created: 2021-05-25T20:28:30.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2021-06-14T08:30:15.000Z (almost 4 years ago)
• Last Synced: 2023-08-31T07:03:44.027Z (over 1 year ago)
• Topics: gpt, gpt-3, gpt-neo, transformers
• Language: Jupyter Notebook
• Homepage:
• Size: 25.4 KB
• Stars: 34
• Watchers: 1
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Basic-UI-Gpt-Neo-low-vram

A basic ui for running gpt neo 2.7B on low vram (3 gb Vram minimum) 


 

Expected speed on pcie-3 with 3gb vram is 0.8s/token or 20s for 25 tokens 


Expected speed on pcie-3 with 8gb vram is 0.4s/token or 10s for 25 tokens 


(with a 2000 token input)