Skip to content
You must be logged in to sponsor Defilan

Become a sponsor to Christopher Maher

@Defilan

Christopher Maher

Defilan
Seattle, WA

LLMKube is a Kubernetes operator for running LLMs on hardware you own: NVIDIA, Apple Silicon, and AMD. It's Apache 2.0 and built in public, including Foreman, a harness that lets local models do real coding work under gates and review (trust the harness, not the model). I develop it on a homelab fleet of consumer GPUs, Apple Silicon, and AMD boxes.

Sponsorship covers the unglamorous parts that keep it moving: GPU electricity, model storage, the eval and benchmark runs, and the time to keep shipping in the open.

Nothing in LLMKube is paywalled and nothing will be. Sponsorship is simply how the open core stays open. If LLMKube saved you a cloud bill, or two hours of debugging vLLM, this is the way to keep it going.

Featured work

  1. defilantech/LLMKube

    Kubernetes operator for self-hosted LLM inference across a heterogeneous GPU fleet: NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal. Runtimes: llama.cpp, vLLM, TGI, mlx-server. Multi-GPU, autoscal…

    Go 144

Learn more about sponsoring developers and organizations.

Report abuse

Select a tier

$ a month

Choose a custom amount.

$5 a month

Select

A GPU-hour. Keeps a card warm. Thank you, genuinely.

$10 a month

Select

An overnight batch. Funds the electricity for a night of local agentic coding.

$25 a month

Select

A model eval. Pays for a model evaluation / benchmark run across the fleet.

$50 a month

Select

Fleet supporter. Everything above, plus your name in BACKERS.md.

$250 a month

Select

Org sponsor. For teams that depend on LLMKube: logo + link in the README. Nothing is paywalled; this funds the roadmap.