Simple framework for LLM inferencing

LLM::Graph can be used to efficiently schedule and combine multiple LLM generation steps for an optimal evaluation managing LLM submissions' concurrency.

WWW::LLaMA provides access to the LLaMA playground.

dosh (do shell) - LLM-powered shell commands.

dosh (do shell) - LLM-powered shell commands.

Raku client for the HuggingFace API — model search and tokenizer downloads

Content classification and model routing for LLM data pipelines

Structured LLM task layer with retry, JSON parsing, and query-based routing

Different subs and CLI scripts for specific but repeatable LLM-based workflows.

MCP (Model Context Protocol) server framework for Raku