search

tairov / llama2.mojo

Inference Llama 2 in one file of pure 🔥
https://www.modular.com/blog/community-spotlight-how-i-built-llama2-by-aydyn-tairov
MIT License
2.09k stars 139 forks source link

Add support for mojo 0.6 #80

Closed clarkezone closed 7 months ago

clarkezone commented 8 months ago

I know it only came out yesterday but ;-)

tairov commented 8 months ago

Hi @clarkezone ! Do you see any issues with running it on Mojo 0.6.0? It looks working fine on my side .

Also do you have any suggestions which new Mojo features you'd like to see implemented in llama2.🔥?

clarkezone commented 8 months ago

Hi @clarkezone ! Do you see any issues with running it on Mojo 0.6.0? It looks working fine on my side .

Also do you have any suggestions which new Mojo features you'd like to see implemented in llama2.🔥?

Just tested if with 0.6 on AMD64 linux and it appears to work fine. So probably safe to bump the readme.

I haven't looked at the llama2.🔥 code yet and still working my way through the 🔥 0.6 readme. So not yet but will do for sure ;-). Nice job with this and getting into the keynote. Now we just need a port of whisper-cpp ;-)