kessler/gemma-gem
Gemma Gem runs Google's Gemma 4 model entirely on-device via WebGPU — no API keys, no cloud, no data leaving your machine.
Language: TypeScript
#ai #chrome_extension #gemma4 #gemma4_2b #gemma4_agent_skills #llm
Stars: 648 Issues: 6 Forks: 60
https://github.com/kessler/gemma-gem

EKKOLearnAI/hermes-web-ui
Web dashboard for Hermes Agent — multi-platform AI chat, session management, scheduled jobs, usage analytics & channel configuration (Telegram, Discord, Slack, WhatsApp)
Language: TypeScript
#ai_agent #chat_ui #chatbot #cron_jobs #dashboard #discord_bot #hermes_agent #i18n #llm #multi_platform #naive_ui #nousresearch #session_management #slack_bot #telegram_bot #typescript #vue3 #web_terminal #whatsapp
Stars: 800 Issues: 12 Forks: 84
https://github.com/EKKOLearnAI/hermes-web-ui

stanford-iris-lab/meta-harness
Reference code for the Meta-Harness paper.
Language: Python
#harness_engineering #llm_agents
Stars: 540 Issues: 1 Forks: 42
https://github.com/stanford-iris-lab/meta-harness

cosmicstack-labs/mercury-agent
Soul-driven AI agent with permission-hardened tools, token budgets, and multi-channel access. Runs 24/7 from CLI or Telegram.
Language: TypeScript
#ai_agent #ai_assistant #llm
Stars: 798 Issues: 6 Forks: 91
https://github.com/cosmicstack-labs/mercury-agent

GammaLabTechnologies/harmonist
Portable AI agent orchestration with mechanical protocol enforcement. 186 agents, zero runtime dependencies.
Language: Python
#agent_framework #agent_system #ai_agents #claude_code #cursor_ide #llm #multi_agent_framework #orchestration #prompt_engineering #python
Stars: 642 Issues: 0 Forks: 273
https://github.com/GammaLabTechnologies/harmonist

future-agi/future-agi
Open-source, end-to-end platform for evaluating, observing, and improving LLM and AI agent applications. Tracing · Evals · Simulations · Datasets · Gateway · Guardrails. Self-hostable. Apache 2.0.
Language: Python
#ai #ai_gateway #evals #llm #observability #simulation
Stars: 662 Issues: 38 Forks: 109
https://github.com/future-agi/future-agi

nexu-io/open-design
🎨 Local-first, open-source alternative to Anthropic's Claude Design. ⚡ 19 Skills · ✨ 71 brand-grade Design Systems · 🖼️ sandboxed preview · 📦 HTML/PDF/PPTX export. 🤖 Runs on Claude Code / Codex / Cursor / Gemini CLI / OpenCode / Qwen.
Language: TypeScript
#ai_agents #ai_design #anthropic #byok #claude #claude_code #claude_design #codex #coding_agents #cursor #design #design_systems #design_tools #gemini_cli #generative_ai #llm #local_first #react #skills #typescript
Stars: 2258 Issues: 20 Forks: 269
https://github.com/nexu-io/open-design

raiyanyahya/how-to-train-your-gpt
Build a modern LLM from scratch. Every line commented. Explained like we are five.
Language: Jupyter Notebook
#attention_mechanism #deep_learning #educational #from_scratch #gpt #language_model #llama #llm #machine_learning #natural_language_processing #python #pytorch #tokenisation #transformers #tutorial
Stars: 478 Issues: 0 Forks: 63
https://github.com/raiyanyahya/how-to-train-your-gpt

strukto-ai/mirage
A Unified Virtual Filesystem For AI Agents
Language: TypeScript
#agent_sandbox #agent_tools #ai_agents #bash #claude_code #fuse #langchain #llm_agents #openai_agents #python #typescript #vfs #virtual_filesystem
Stars: 1136 Issues: 6 Forks: 69
https://github.com/strukto-ai/mirage

lightseekorg/tokenspeed
TokenSpeed is a speed-of-light LLM inference engine.
Language: Python
#blackwell #deepseek #gpt_oss #kimi #lightseek #llm #minimax #qwen #speed_of_light #tokenspeed
Stars: 695 Issues: 15 Forks: 43
https://github.com/lightseekorg/tokenspeed