[Doc]: Compare LMDeploy vs vLLM AWQ Triton kernels

[casper-hansen]

📚 The doc issue

LMDeploy has their own Triton kernels implemented for AWQ. It would be interesting to know which Triton kernels are more efficient and if vLLM can improve the performance.

Link to kernels: https://github.com/InternLM/lmdeploy/blob/0c80baa001e79d0b7d182b8a670190801d2d8d5b/lmdeploy/pytorch/kernels/cuda/awq_kernels.py

Suggest a potential alternative/fix

No response

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.