dongmao zhang

this plugin is really great!. but I think it still needs some enhancement. 1. source page and preview page could scroll synchronously. 2. maybe this tool needs more export format...

mistral does not support NVIDIA V100 (compute_cap <= 800)

1

**Describe the bug** it does not support some old hardware. Can it just convert bfloat16 to float16 before loading model. just like vllm is doing?

Add bitsandbytes fp4 support

1

bitsandbytes could support read bnb fp4 model such as PrunaAI/Einstein-v6.1-Llama3-8B-bnb-4bit-smashed. I can not find any model which is tiny and fast. So I did not add new unit test. this...

thank you for this great project. I think it helps a lot for RDMA beginner. But if I want to copy gpu memory to remote cpu memory thought RDMA. is...

[Feature] Proposal for adding PD-Disaggregation Feature to SGLang

3

## Principles of Design 1. Model-Agnostic Approach The changes will be implemented in a model-agnostic manner, rather than being tailored to specific models. 2. Compatibility with Open-Source Projects To enhance...