#Neuro python migration?
1 messages · Page 1 of 1 (latest)
Using it for inference straight up requires minimal changes on code. There is a waiting list so best signup early (I still hasn't got reply). IMHO pytorch/huggingface pipeline will probably fork this straight up so zero code changes will be required eventually.
Update: So i got into the beta program, the bad news is they will only provide you access to a Jupyter Notebook where it is pre-installed.
None of the existing pytorch or huggingface library has been ported, but it got some very fast matrix math and it seems to use gpu by default, no CUDA required.
Mojo is a new LLVM programming language designed as a superset of Python with the low-level performance of C. It is optimized to run on GPUs with CUDA and other exotic hardware for deep learning and Artificial Intelligence.
#programming #tech #thecodereport
💬 Chat with Me on Discord
🔗 Resources
- Mojo Language...