Open-source projects tagged “llm”

Hermes Agent skills that have the agent write its own analytical prompts (prisms) tailored to each artifact, then report both findings and blind spots.

An agent skill that makes AI coding agents report task results in controlled Simplified Chinese, based on ASD-STE100 principles.

A self-improvement add-on for Hermes Agent that analyzes session failures, patches weak skills, runs GEPA evolution, and reports progress.

Local inference engine that runs the large Qwen3.8-Flash-Next model on consumer NVIDIA or AMD GPUs under Windows or Linux, with OpenAI and Anthropic compatible APIs.

An agentic coding tool from Anthropic that runs in your terminal, reads your codebase, and handles coding tasks and git workflows via natural language.

A collection of starter projects for building deployable applications on the Claude API, covering agents, computer use, browser automation and finance analysis.

A large collection of role-specific AI agent prompt files (engineering, design, marketing, sales, security and more) that install into tools like Claude Code, Cursor and Gemini CLI.

Turns a pasted URL or a text prompt into a 20-40 second launch video, written as HTML by Claude Opus and rendered to MP4 by a serverless agent.

A structured, reusable prompt design framework for LLMs, with a template of Role, Profile, Goal, Skills, Rules and Workflow sections.

NEO is a family of native vision-language models that process pixels and words in one encoder-free dense architecture, with training and evaluation code.

Open-source Python package that turns websites, PDFs, images, YouTube videos or topics into multilingual audio podcast conversations using GenAI.

Open-source cross-platform desktop app that generates short marketing videos from text prompts and footage, with AI copywriting, TTS voiceover and automatic batch editing.

Unity Editor and runtime plugin and MCP server that lets AI agents like Claude, Cursor and Copilot operate Unity projects and run LLMs inside games.

Unreal Engine plugin that connects games and editor tooling to LLM, image, TTS and 3D generation APIs, plus an MCP server for AI assistants.

A unified training engine for multimodal models, supporting VLM, diffusion and LLM training with FSDP2, Muon and Liger kernels.

A collection of remote MCP servers that let AI clients like Cursor or Claude query and manage Cloudflare accounts and services in natural language.

A Go command-line tool and web UI for downloading models and datasets from the HuggingFace Hub with parallel connections and resume support.

Browser automation platform that turns recorded browser actions into MCP commands or Claude Skills so AI agents can control browsers.

A self-hosted personal AI assistant that runs on your own devices and answers you through messaging apps like WhatsApp, Telegram, Slack and Discord.

An MCP server that exposes Ghidra's reverse engineering functions so LLM clients can decompile and analyze binaries.