vLLM · paid microVM · no free trial
vLLM

vLLM agent templates you can fork. Your keys. Your merge button.

High-throughput LLM serving. Stand up an OpenAI-compatible endpoint and measure what it actually does under load.

If you already use vLLM, start under Fork these now. A fork boots a paid Firecracker microVM from $25/mo — BYOK, not a shared container. First purchase: 30-day money-back. Rows labeled "Not forkable yet" have no button. Operators draft; you still read the code. Nothing auto-merges or auto-deploys.

What happens next: open a ready operator → Fork → checkout → microVM provisions → run the demo prompt on your box. 1 forkable operator listed below.

All Software Factory toolsAll collections1 forkable · 1 listed · $25+/mo · 30-day money-back

Get eng operators in your inbox

GitHub, Linear, Sentry, CI — short notes when a ready operator is worth opening. Drafts only: you still read the diff and merge. Browse either way. Unsubscribe anytime.

Fork these now

Founder-tested operators. Fork, paste the demo prompt, get a real output. Your keys stay on your microVM.

  • OpenScience logo

    vLLM Inference Lab: p99 Bench

    OpenScience

    $49/mo

    OpenScience lab for OpenAI-compatible inference: design a vLLM-style serve stack, bench it, read real p50/p99. BYOK Anthropic plus optional HF token.

    00Open · $49/mo

Ready operator → Fork → paid microVM → demo prompt. First purchase is covered by a 30-day money-back guarantee. No free trial. Prefer another tool? Back to Software Factory shelves.