ifsheldon

> However, we want to keep things simple without falling into overengineering trap, above all, just working with openai should be prioritized even if that means library doesn't work with...

@64bit Why did you want `Config: Clone` in the first place? I looked through the codebase, it seems `config.clone` is never used. So, after removing the trait bound `Clone` in...

> Is that enough or are there additional changes needed? For Azure OpenAI services, we need `API_KEY` and more, which are `deployment_id` and `endpoint`

I encountered this as well. But somehow this PR is still not merged.

Thanks! Awesome! These early results seem to have IPAdpater-like capabilities. Probably it also has strong in-context learning capability like given a pair of (original image, target image) as an example,...

> I reproduced a version of Ella based on SD XL, and the effect is really improved over the original SD XL. I have read your work on multimodal integration,...

+1. I'm doing quantum computing simulation on my Mac with PyTorch, but PyTorch has incomplete and slow MPS support. I'm considering implementing quantum simulation algorithms in MLX, but I need...

As per [my comment](https://github.com/zed-industries/zed/issues/18891#issuecomment-2466595547), I can build a `remote-server` for riscv64 with a couple of dependency bumps. The toolchain I used to compile it on my RISC-V dev board is...

OK, thanks for the clarification! > What kind of machines run risc64? If there's a way to boot up a reasonably quick one in the cloud, we could potentially publish...

I got the same/very similar issue when I tried to compile zellij from source on my RISC-V board.