turboderp-org/exllamav2
1
0
Fork 0
mirror of https://github.com/turboderp-org/exllamav2.git synced 2026-04-28 18:21:36 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fixed minor typo in convert.md doc (#463)

Browse Source 
changed '64 GB or RAM' to '64 GB of RAM'
This commit is contained in:
rohitanshu
2024-05-27 02:56:17 +05:30
committed by GitHub
parent 5ef9b13d88
commit b5f110d215
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
doc/convert.md
View File

@@ -149,7 +149,7 @@ catastrophically.
### Hardware requirements
Roughly speaking, you'll need about 64 GB or RAM and 24 GB of VRAM to convert a 70B model, while 7B seems to require
Roughly speaking, you'll need about 64 GB of RAM and 24 GB of VRAM to convert a 70B model, while 7B seems to require
about 16 GB of RAM and about 8 GB of VRAM.
The deciding factor for the memory requirement is the *width* of the model rather than the depth, so 120B models that