Alex Cheema
Alex Cheema
> we are loading up mac studios and notice that on subsequent inference, memory will suddenly spike on one of the nodes as if it is trying to reload model...
Try again on latest version. Some fixes in there for Linux, specifically around memory usage
Merged. Please email [email protected] with your ethereum address for your $200 bounty in USDC.
Are you using a terminal emulator? Which device? Which version of android?
The small llama 3.2 language models are already supported. The larger multimodal vision language models are yet to be implemented: #247
> When using the LLama 3.3 70B model, the text printed in the terminal is normal, but the content returned by the interface shows garbled Chinese characters. Hi, sorry to...
Hold off on this until we have PyTorch support merged (#139). You could already try, but others have run into issues and I'm not sure of the root cause.
I'm not sure about this. In tinygrad we use the default device which should be CUDA on an NVIDIA GPU instance. Source: https://docs.tinygrad.org/mnist/
Closing as I don't think this is necessary. Typically this means you didn't install CUDA.