DEV Community

Pujun Lun profile picture

Pujun Lun

404 bio not found

When "Just Enable the CUDA Feature" Met Two Free GPUs

When "Just Enable the CUDA Feature" Met Two Free GPUs

Comments
7 min read

Want to connect with Pujun Lun?

Create an account to connect with Pujun Lun. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
What I Gave Up to Make Speculative Decoding Faster

What I Gave Up to Make Speculative Decoding Faster

1
Comments
5 min read
When LLM Inference on CPU Beats Waiting for the GPU

When LLM Inference on CPU Beats Waiting for the GPU

1
Comments 4
5 min read
Why My Prefix Cache Missed Identical Requests

Why My Prefix Cache Missed Identical Requests

1
Comments 3
6 min read
Why Continuous Batching Made My LLM Inference Engine Slower

Why Continuous Batching Made My LLM Inference Engine Slower

Comments 1
4 min read
From a Rust Inference Engine to Heterogeneous Serving

From a Rust Inference Engine to Heterogeneous Serving

Comments 1
3 min read
loading...