Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

invalid model file './models/llama-13b-4bit.pt' (bad magic) #81

Closed
acheong08 opened this issue Mar 13, 2023 · 3 comments
Closed

invalid model file './models/llama-13b-4bit.pt' (bad magic) #81

acheong08 opened this issue Mar 13, 2023 · 3 comments
Labels
model Model specific

Comments

@acheong08
Copy link

acheong08 commented Mar 13, 2023 •
edited
Loading

Downloaded from

magnet:?xt=urn:btih:36945b5958b907b3ab69e963ba0de1abdf48c16c&dn=LLaMA-HFv2-4bit&tr=http%3a%2f%2fbt1.archive.org%3a6969%2fannounce&tr=http%3a%2f%2fbt2.archive.org%3a6969%2fannounce
@thement
Copy link
Contributor

thement commented Mar 13, 2023

Downloaded from

magnet:?xt=urn:btih:36945b5958b907b3ab69e963ba0de1abdf48c16c&dn=LLaMA-HFv2-4bit&tr=http%3a%2f%2fbt1.archive.org%3a6969%2fannounce&tr=http%3a%2f%2fbt2.archive.org%3a6969%2fannounce

Those are weights for torch library. Llama.cpp uses different format and afaik it is not loadable yet.

@gjmulder
Copy link
Collaborator

You need to run python3 convert-pth-to-ggml.py and quantize on the torrent *.pth files to convert them into the correct format as per the instructions in the README.md

@acheong08
Copy link
Author

I just downloaded the original files and quantized them. Worked well

@acheong08 acheong08 closed this as not planned Won't fix, can't repro, duplicate, stale Mar 13, 2023
@gjmulder gjmulder added the model Model specific label Mar 15, 2023
Hades32 pushed a commit to Hades32/llama.cpp that referenced this issue Mar 21, 2023
@antimatter15
Automatically Generate Releases for Mac, Linux, Windows (ggml-org#81)
99f3908 
* trying to build for arm

* cross compiling?

* only mac

* gh actions

* fix

* fixing dash

* ensure that its executable

* adding back windows

* unindent

* typo

* oops

* runs on

* more release stuff

* more release stuff

* fix workflow

* adding back linux builds

* conditional stuff
@Bearsaerker Bearsaerker mentioned this issue Mar 12, 2025
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
model Model specific
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@gjmulder @acheong08 @thement