ifsheldon

Results 77 issues of ifsheldon

### 💻 系统环境 | Operating System macOS ### 🌐 浏览器 | Browser Safari ### 🐛 问题描述 | Bug Description I have a public domain, on which I hosted a GPT...

🐛 Bug

Hi! Awesome work! It's super nice that you provided a docker image. I am wondering if we run StreamDiffusion in a docker container, it seems we can't run the screen...

Can we support multi-GPU inference? The model is too large to fit in one consumer GPU, even 4090 does not have enough memory. I think two 4090s are enough.

### 🥰 Feature Description 支持 MiniMax 的模型,MiniMax 的模型应该是国内模型和 OpenAI 最接近的了,无论是性能还是 API ### 🧐 Proposed Solution 没有官方库,所以只能用 http 请求;或者 one-api 支持 MiniMax 之后可以用 one-api,见 https://github.com/songquanpeng/one-api/pull/413 ### 📝 Additional Information _No response_

Inactive
🌠 Feature Request

### Self Checks - [X] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [X] I confirm that I am using English to file this...

bug
enhancement

Hey! Here is just a heads-up that `wgpu` 0.18 and `winit` 0.29.2 came up recently with a lot breaking changes. For the breaking changes from `wgpu`, please see https://github.com/gfx-rs/wgpu/releases/tag/v0.18.0. I...

Hi! I've been trying to porting nanoGPT to Rust with dfdx. The `transformer` module is awesome! but it seems an important trick is missing, which is the attention mask in...

new feature

Hi! I read your training do. As far as I understand, without hacking the code, we now can only use or implement filters to generate control images on the fly?...

This is a minimum code example for incrementally printing markdown. This should close #42

Thanks to GPT-4, I solved the compile error of example render-input-markdown on my Mac. There is an alternative method, which requires a `build.rs`, in the solutions given by GPT-4, but...