Skip to content

viennh/CAP6614-PowerInfer

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1,598 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PowerInfer: Consumer-GPU LLM Serving (UCF project fork)

This repository is a fork of Tiiny-AI/PowerInfer for our UCF CAP 6614 project work. For full upstream build instructions, model support, and engine documentation, see README-PowerInfer.md in this repo, and the original projects linked below.

Project team

Name Email
Cuong Dang cuong.dang@ucf.edu
Hai Nguyen hai.nguyen@ucf.edu
Joshua Lowe jo201760@ucf.edu
Lam Nguyen lam.nguyen@ucf.edu

System paper: PowerInfer

Title PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU
Authors Xue, Y., et al. (SJTU IPADS)
Venue SOSP 2024

Short description: PowerInfer exploits locality in LLM inference: for any given input, only a small set of neurons are hot (frequently activated). The system pre-loads hot neurons on the GPU while computing cold neurons on the CPU, reducing memory pressure and data movement. Reported results include up to about 11× speedup over llama.cpp on a single NVIDIA RTX 4090 for large models (e.g. 175B-class).

Upstream code (reference): github.com/SJTU-IPADS/PowerInfer

Our project angle

We deploy PowerInfer on a consumer GPU with an OPT or LLaMA-family model, then:

  1. Profile the hot/cold neuron distribution across layers and runs.
  2. Compare end-to-end inference speed against llama.cpp and vLLM under controlled settings.
  3. Analyze how the hot-neuron set varies with different input types (e.g. topic, length, or task format).

Project artifacts (scripts, plots, and result tables) live under Reports/ in this repository.

Experimental setup

Full procedure (prerequisites, UCF Newton, Apple rebuild options, and Phases 1–4) is in Reports/ExperimentalSetup.md. Guidelines in brief:

  • Where it runs best: The paper’s GPU–CPU hybrid (hot on GPU, cold on CPU) is built for NVIDIA CUDA. On Apple Silicon, runs are CPU-oriented (e.g. NEON/BLAS, not the discrete-GPU hot/cold path). GPU benchmarks and apples-to-apples comparisons with that design belong on a CUDA machine—this project uses UCF Newton; profiling and much analysis can be done locally.
  • ~16 GB VRAM nodes: A single F16 dense run is most realistic for ~7B; 13B/70B often need a larger-GPU partition, quantized baselines (e.g. llama.cpp), or you document partial-GPU/CPU behavior. Phase 2 neuron profiling is CPU and disk only, not a VRAM bottleneck.
  • HPC env (Newton): Typical bootstrap is module load an Anaconda module, a project .venv, and pip install -r requirements.txt—see the report for exact module names and run scripts under Reports/scripts/ucf_newton/.

Links

Resource URL
This fork (base) Tiiny-AI/PowerInfer
Original PowerInfer (SJTU) SJTU-IPADS/PowerInfer

License and attribution

This project inherits the upstream project’s terms; see the license files in the tree and README-PowerInfer.md for full notices. When you cite the system in academic work, use the SOSP 2024 PowerInfer paper and the upstream repository as appropriate to your style guide.

About

High-speed Large Language Model Serving for Local Deployment

Resources

License

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • C++ 56.1%
  • C 16.0%
  • Python 9.3%
  • Cuda 6.9%
  • Objective-C 2.3%
  • Metal 2.2%
  • Other 7.2%