Skip to content
This repository was archived by the owner on Feb 25, 2026. It is now read-only.

pestopoppa/amd-epyc-inference

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

674 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

amd-epyc-inference (Archived)

This repository has been split into focused repos. See epyc-root for setup instructions.

New repositories

Repository Purpose
epyc-root Umbrella repo — governance, coordination, setup scripts
epyc-orchestrator Production multi-model orchestration system
epyc-inference-research Benchmarks, experiments, model evaluation
epyc-llama Custom llama.cpp fork with AMD EPYC patches

Quick start (new setup)

git clone https://github.com/pestopoppa/epyc-root.git
cd epyc-root
./scripts/setup.sh

Why the split?

This monorepo mixed four distinct concerns: production orchestration, inference research, llama.cpp patches, and cross-repo governance. Splitting improves:

  • FOSS accessibility — orchestrator can be used without research/governance baggage
  • Focused development — each repo has its own CLAUDE.md, tests, and conventions
  • Dependency clarity — formal dependency map in epyc-root

Historical reference

This repository is preserved as-is for historical reference. All commit history remains intact. For active development, use the repositories listed above.

About

LLM inference optimization on AMD EPYC 9655 Turin - speculative decoding, MoE optimization, llama.cpp benchmarks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages