N
One option is to use the `bump` allocator for now.
there is a check for [is_local](https://github.com/alloy-rs/alloy/blob/57dd4c538293421c3d1a793cba79ad6f46d6444f/crates/rpc-client/src/client.rs#L182), maybe there could be a setter at the provider level but i don't think this is a bug
Created a fork here that I plan on maintaining. I bumped all the external integrations to the most recent versions. https://github.com/nhtyy/deepsize2
@pen-pal Hi we currently only support SM86 and SM89, you can find the full list here: https://developer.nvidia.com/cuda-gpus
Closing for now as stale