Open-source LLM apps and frameworks

GarethManning

education-agent-skills

★ 978

165 evidence-grounded AI skills for teachers, school leaders and EdTech builders—pedagogy, learning science, curriculum, assessment and regeneration. Claude, Codex and Hermes.

DrEAmSs59

CS2-insight-agent

★ 977

An all-in-one CS2 content creation suite: demo library and analysis (highlights, lowlights, mock-pro POV, 2D replay, heatmaps, skin swapping), automatic OBS recording, a compilation workbench with LiteCut multi-track editing, and optional snarky LLM commentary. Zero injection, zero hooking, zero reverse-engineering of the game process -- low-risk local replay only.

microsoft

wassette

★ 976

Wassette: A security-oriented runtime that runs WebAssembly Components via MCP

OpenSparX

MasterAgent

★ 976

Build AI agents that run 100% on-device. Sub-100ms latency on Qualcomm NPU. Zero cloud dependency.

jongpie

NebulaLogger

★ 975

The most robust observability solution for Salesforce experts. Built 100% natively on the platform, and designed to work seamlessly with Apex, Lightning Components, Flow, OmniStudio, and integrations.

salesforce

OmniXAI

★ 975

OmniXAI: A Library for eXplainable AI

Vrun-design

openflowkit

★ 975

100% Free, Open-source local-first AI diagramming for architecture diagrams and flowcharts with animated exports.

alikon-art

DeterminFlow

★ 974

A production-oriented AI workflow runtime for building, validating, recovering, and shipping complex AI workflows as dependable services.

e10nMa2k

cc-mini

★ 974

Ultra-light Harness scaffolding for AI agents, a mini version of claude code

friuns2

codex-mobile

★ 974

🚀 Run Codex Mobile Anywhere: Linux, Windows, or Termux on Android 🚀

luongnv89

asm

★ 974

The universal skill manager for AI coding agents.

kessler

gemma-gem

★ 973

Gemma Gem runs Google's Gemma 4 model entirely on-device via WebGPU — no API keys, no cloud, no data leaving your machine.

TryCaspian

caspian-sdk

★ 973

Agent communication SDK. The open-source agent communication layer for AI agents — email, WhatsApp, Slack, Discord, Telegram, SMS. Python & TypeScript.

XYZ-AI-Lab

AxisAgentic

★ 973

AxisAgentic: An Extensible Runtime and Trajectory-Collection Framework for Long-Horizon Agents.

breschio

drawbridge

★ 972

Design editor for Claude Code and Cursor. "Figma Comments" for the browser, are sent directly to Claude Code and Cursor as prompts.

freestylefly

wesight

★ 972

Open-source desktop AI agent workspace with one-click Claude Code, Codex, OpenClaw, Hermes Agent setup and custom LLM model routing.

Kulaxyz

self-learning-skills

★ 972

A self-improving skill for AI coding agents (Claude Code, Cursor, AGENTS.md): recognize a hard-won golden path in a session and harvest it into a reusable skill/rule for next time.

Electricitysheep

dsh-handbook

★ 971

A from-zero-to-one handbook for DeepSeek Harness (dsh): installation, plugin development, performance tuning, real-world case studies, and multi-agent comparisons on the same model (Chinese + English PDF).

lich0821

ccNexus

★ 971

Intelligent API gateway for Claude Code and Codex CLI - rotate endpoints, monitor usage, and seamlessly integrate OpenAI, Gemini, and other platforms.

china-qijizhifeng

agentic-harness-engineering

★ 970

Official AHE code — Agentic Harness Engineering: observability-driven automatic evolution of coding-agent harnesses (concurrent w/ meta-harness). NexAU-AHE reaches 84.7% ± 2.1 pass@1 on Terminal-Bench 2 (GPT-5.5). Lifts GPT-5.4 69.7→77.0% over 10 iters, beats Codex/ACE/Training-Free GRPO; frozen harness transfers to SWE-bench-Verified.

data-prep-kit

data-prep-kit

★ 969

Open source project for data preparation for GenAI applications

mbzuai-oryx

groundingLMM

★ 969

[CVPR 2024 🔥] Grounding Large Multimodal Model (GLaMM), the first-of-its-kind model capable of generating natural language responses that are seamlessly integrated with object segmentation masks.

openeverest

openeverest

★ 969

OpenEverest is an open-source platform for automated database, storage and LLMs provisioning and management. It supports multiple technologies and can be hosted on any Kubernetes infrastructure, in the cloud or on-premises.

AgileRL

AgileRL

★ 968

Streamlining reinforcement learning with RLOps. State-of-the-art RL algorithms and tools, with 10x faster training through evolutionary hyperparameter optimization.