Just a package

MCP server for Unsloth - a library for optimizing and fine-tuning LLMs

Agent skill for managing LLM fine-tuning through Unsloth Studio API

Fine-tune cheap open-source LLMs (GLM, Qwen, DeepSeek) on your AI coding agent's successful runs with LoRA (SFT + DPO) so your model cascade escalates to expensive frontier models (GPT, Claude) less often — cutting cost-per-resolved. Turns run history int

Bundled LLM tokenizer vocabularies (tokenizer.json set) for @plurnk/plurnk-mimetypes' tokenizer seam — exact token counting for window math.

A private reasoner for your terminal. Direct conversation or grounded multi-agent research, GPU-native and fully local. No API keys, no inference servers.

TUI configurator and local API proxy for wiring custom model providers into OpenCode and Claude Code.

gitless downloader for Hugging Face repositories

Telegram bridge for Pi local coding agents: terminal-continuation and business manager modes

Background task queue for OpenCode projects, powered by TASKS.md

Makora provider plugin for OMP — Access DeepSeek V4, GLM 5.1, Kimi K2.6, Llama 3.3, Qwen 3.6, and more through the Makora inference API

Cursor on an instant-noodle budget 🍜 — use Antigravity, Codex, z.ai, MiniMax, Opencode, LMStudio & more for free or cheap

Generative Art on HTML Canvas

Code-first TypeScript training SDK, CLI, and local Studio.

Chutes Models Plugin for OpenCode - Access 58+ state-of-the-art AI models through the Chutes API

Bootstrap provider-aware AI context files for projects with extensible LLM plugins

Tree-sitter indexed lookups — smart code reading for AI agents

Lemon AI Agent — spawn specialist agents from Node.js APIs with tools, streaming, and structured output

Simple terminal-based translation tool

Self-contained local LLM coding agent CLI powered by Qwen3-8B