* [opencodex v2.83.0](https://github.com/lidge-jun/opencodex) – Lightweight local proxy that translates Codex Responses API into multiple LLM providers with full support for streaming, tools, reasoning tokens, and images.

* [Reasonix v2.34.0](https://github.com/esengine/DeepSeek-Reasonix) – DeepSeek-native AI coding agent for your terminal, engineered around prefix-cache stability to keep token costs low across long sessions.

* [CopilotKit v1.78.0](https://github.com/CopilotKit/CopilotKit) – Tool for building integrated AI assistants and agents that operate within user applications.

* [Langfuse v4.56.0](https://github.com/langfuse/langfuse) – LLM engineering platform for collaboratively developing, monitoring, evaluating, and debugging AI applications.

* [Token-Oriented Object Notation (TOON) v4.4.0](https://github.com/toon-format/toon) – Compact, human-readable serialization format optimized for structured data in large language model inputs with reduced token usage.

* [Agent Canvas v1.26.0](https://github.com/OpenHands/OpenHands) – Self-hosted developer control center that runs coding agents across local, remote, or cloud backends and automates everyday engineering workflows.

* [Ollama v0.40.0](https://github.com/ollama/ollama) – Tool for running and managing large language models.

* [Ax 25.3.0](https://github.com/ax-llm/ax) – End-to-end streaming framework for building multi-modal agents with typed signatures.

* [SigMap v8.74.0](https://github.com/manojmallick/sigmap) – Finds relevant files in a codebase and supplies compact function and class signatures to AI models to reduce tokens and improve answers.

* [Context7 MCP mcpb-v4.3.0](https://github.com/upstash/context7) – Fetches up-to-date documentation and code examples for prompts in LLMs.

* [FastGPT v4.17.2](https://github.com/labring/FastGPT) – Knowledge-based platform leveraging LLMs for data processing, retrieval, and AI workflow orchestration.

* [LangChain langchain-openai==1....](https://github.com/langchain-ai/langchain) – Framework for building agents and LLM-powered applications with modular components and integrations.

* [The Web MCP v2.12.0](https://github.com/brightdata/brightdata-mcp) – MCP-based server that provides LLMs real-time, unblocked access to live web data and browser automation.

* [Token Monitor v0.68.0](https://github.com/Javis603/token-monitor) – Desktop widget that monitors live token usage and AI tool limits across multiple AI coding tools with real-time multi-device sync and usage history breakdowns.

* [RubyLLM v2.1.0](https://github.com/crmne/ruby_llm) – A Ruby library providing a unified API for interacting with various AI models and services.

* [Token Optimizer MCP v7.4.0](https://github.com/ooples/token-optimizer-mcp) – Model Context Protocol server that optimizes Claude Code context usage using caching, Brotli compression, and smart tool replacements for large text and file operations.

* [easy-stock v1.5.0](https://github.com/jundizhou/easy-stock) – Local-first A 股行情分析与AI智能投研代理，支持日常共识与短线研判。

* [Frontman v5.3.0](https://github.com/frontman-ai/frontman) – Browser-based tool that edits a project's frontend source code from live UI interactions and natural-language instructions.

* [obsidian-mcp-server v3.7.0](https://github.com/cyanheads/obsidian-mcp-server) – MCP server for Obsidian vaults that reads, writes, searches, and surgically edits notes, tags, and frontmatter via the Local REST API plugin.

* [Promptfoo 0.124.0](https://github.com/promptfoo/promptfoo) – Developer-friendly local tool for testing, evaluating, and securing large language model applications.

* [LLM Gateway v1.20.0](https://github.com/theopenco/llmgateway) – API gateway that routes and manages requests across multiple LLM providers with an OpenAI-compatible API, usage analytics, and cost tracking.

* [avoid-ai-writing v3.37.0](https://github.com/conorbronsdon/avoid-ai-writing) – Audit and rewrite content to remove AI writing patterns, with detect-only or edit-in-place modes and configurable voice profiles for agent skills.

* [HuggingFaceModelDownloader v3.4.0](https://github.com/bodaay/HuggingFaceModelDownloader) – Fast, resilient, resumable CLI and Go library for downloading models and datasets from the Hugging Face Hub with live TUI and JSON events.

* [Cherry Studio v2.1.6](https://github.com/CherryHQ/cherry-studio) – Desktop client supporting multiple LLM providers, available on Windows, Mac, and Linux.

* [Hippo v1.71.0](https://github.com/kitfunso/hippo-memory) – Biologically inspired memory layer for AI agents with decay, consolidation, and cross-tool portability.

* [Coddy 1.3.0](https://github.com/coddy-project/coddy-agent) – Coddy is a distroless-friendly Go-based general-purpose agent harness with a ReAct loop, ACP/HTTP/Telegram interfaces, MCP and skills integration, and long-term memory.

* [aimock v1.45.0](https://github.com/CopilotKit/aimock) – Mock infrastructure for testing AI applications' external services like LLMs, multimedia APIs, agent protocols, and vector databases.

* [Bloks v2.6.0](https://github.com/hamedgitty/bloks) – Local-first desktop workspace that runs personal AI agents in rooms with chained leadership, local memory, file-based skills, and human-gated actions.

* [AI Reader V2 v0.82.0](https://github.com/mouseart2025/AI-Reader-V2) – Novel-analysis visualization tool producing character relationship graphs, maps, timelines and encyclopedias with local and cloud LLM support.

* [Read Frog v1.50.3](https://github.com/mengxi-ream/read-frog) – AI-powered browser extension for immersive language learning with translation and article analysis.

* [Open Multi-Agent v1.21.2](https://github.com/open-multi-agent/open-multi-agent) – TypeScript-native multi-agent orchestration that decomposes goals into parallel task DAGs and synthesizes results.

* [Toonflow v2.0.5](https://github.com/HBAI-Ltd/Toonflow-app) – AI app that converts stories and scripts into animated short dramas with scriptwriting, storyboarding, and character/video generation.

* [RubyLLM::Agents v3.16.0](https://github.com/adham90/ruby_llm-agents) – Rails engine providing an agent DSL, execution tracking, cost analytics, reliability controls, and a real-time dashboard for LLM-powered agents.

* [TencentDB Agent Memory v2.0.2](https://github.com/TencentCloud/TencentDB-Agent-Memory) – TencentDB Agent Memory provides symbolic short-term and layered long-term memory for AI agents to retain context and reduce token usage without external APIs.

* [Prompt Optimizer v2.11.11](https://github.com/linshenkx/prompt-optimizer) – Tool for enhancing prompt quality to improve AI interaction results.

* [MCP PostgreSQL Operations Server 3.4.0](https://github.com/call518/MCP-PostgreSQL-Ops) – Professional MCP server for PostgreSQL database operations, monitoring, and performance analysis with version-adaptive features.

* [Output v0.15.0](https://github.com/growthxai/output) – TypeScript framework for building and orchestrating AI workflows and agent-driven development with integrated prompts, tracing, evals, and secrets management.

* [CortexDB v2.124.0](https://github.com/liliang-cn/cortexdb) – Pure-Go, single-file embedded AI memory and knowledge graph library that provides hybrid vector and lexical retrieval, RDF/SPARQL graph features, and MCP tools via SQLite.

* [GNO v2.10.0](https://github.com/gmickel/gno) – Local knowledge engine for notes, code, PDFs, and Office docs with hybrid search, citations, and verified, explainable answers via CLI, Web UI, REST API, SDK, and MCP.

* [OpenWorkBuddy v0.12.0](https://github.com/CatCatUncle/openworkbuddy) – Local-first AI office agent that generates and edits PPT, Word, and Excel files by using pluggable LLM backends without storing chat logs.

* [Mastra @mastra/core@1.75.0](https://github.com/mastra-ai/mastra) – Opinionated TypeScript framework for building AI applications and features quickly.

* [Open Executive v0.5.2](https://github.com/SenteLabsAI/OpenExecutive) – Highly configurable AI virtual executive persona with eight specialist agents, RAG from knowledge and company documents, episodic memory, and workflow scheduling across common chat and email tools.

* [yzma v1.29.0](https://github.com/hybridgroup/yzma) – Go-based library for hardware-accelerated local inference with llama.cpp integration.

* [yibiao-simple v2.30.14](https://github.com/FB208/OpenBidKit_Yibiao) – Out-of-the-box AI bid writing and checking tool with knowledge base and duplicate detection.

* [FindJobs-Agent v1.1.0](https://github.com/he-yufeng/FindJobs-Agent) – Full-stack job search assistant that crawls postings, uses LLMs to analyze skills, matches them to parsed resumes, and runs AI mock interviews.

* [NobodyWho nobodywho-kotlin-v5....](https://github.com/nobodywho-ooo/nobodywho) – Inference engine for running LLMs locally and efficiently on any device.

* [Sokuji v0.43.0](https://github.com/kizuna-ai-lab/sokuji) – Desktop application providing live speech translation and audio routing using multiple AI-powered APIs.

* [PrismOS-AI v0.7.0](https://github.com/mkbhardwas12/prismos-ai) – PrismOS-AI is a desktop AI that answers questions from local Ollama models, stores knowledge on disk, and works offline with a multi-agent pipeline.

* [Vellium v1.8.0](https://github.com/tg-prplx/vellium) – Desktop chat app using Electron, local Express API, and SQLite for roleplay and creative-writing LLM workflows.

* [jev4pg v0.7.0](https://github.com/Sheltercosmo/jev4pg) – Semantic database for PostgreSQL that converts natural language into SQL and runs 41 JEV operators for filtering, extraction, ranking, and verification.

* [chatless AI v0.7.0](https://github.com/kamjin3086/chatless) – Lightweight local AI chat client built with Tauri 2.0 and Next.js supporting multiple AI APIs.

* [three.ws forge-v0.5.0](https://github.com/nirholas/three.ws) – Browser-native 3D AI agent platform that turns prompts or GLB inputs into textured models with LLM brains and on-chain registration.

* [Ollama GUI v2.0.2](https://github.com/HelgeSverre/ollama-gui) – Single-page web interface for chatting with local Ollama models directly from the browser, with streaming responses, attachments, and local chat history.

* [Fora v0.0.10](https://github.com/imfeh2/fora) – Organization layer for running your own organization of agents.

* [Ekko Studio v0.7.33](https://github.com/EKKOLearnAI/ekko-studio) – Local-first AI workspace providing multi-agent chat, coding, and visual workflow automation across desktop and self-hosted web.

* [Kortix v0.13.53](https://github.com/kortix-ai/suna) – Autonomous company OS where Linux-sandboxed AI agents run code, manage files, and automate operations 24/7.

* [AIKit v0.23.2](https://github.com/kaito-project/aikit) – Platform to host, deploy, build, and fine-tune large language models with OpenAI API compatibility.

* [NoteGen note-gen-v0.38.2](https://github.com/codexu/note-gen) – Cross-platform Markdown note-taking application using AI to organize fragmented knowledge into readable notes.

* [Pinvou Agent v0.10.7](https://github.com/Pinvou/pinvou-agent) – Desktop AI agent workspace that connects models with tools and personal knowledge to produce and manage real deliverable artifacts.

* [Pixeltable v0.7.16](https://github.com/pixeltable/pixeltable) – Pixeltable Cloud provides a unified multimodal backend for AI data apps, combining database, orchestration, and serving in a single Python file.

* [JanusX v0.9.0](https://github.com/TreeX-X/JanusX) – AI coding workspace with parallel Git worktrees, split terminals for multiple agents, and checkpoint-based diff restore with Blueprint and Island workflows.

* [MCP Mesh v4.458.0](https://github.com/decocms/studio) – Control plane routing MCP client traffic through one governed endpoint with auth, RBAC, policy enforcement, and observability.

* [NodeTool v0.8.2](https://github.com/nodetool-ai/nodetool) – Visual platform for building and deploying AI workflows, agents, and multimodal pipelines via drag-and-drop nodes.

* [QQ Agent Plus v0.8.0](https://github.com/sakurawwwxh/qq-agent-plus) – Linux QQ group chat agent for OneBot v11 with segmented messages, sticker support, proactive sending, and built-in operations commands.

* [Nerve v1.5.4](https://github.com/daggerhashimoto/openclaw-nerve) – Self-hosted web cockpit for OpenClaw agents with voice, live workspace editing, charts, scheduling, and token visibility.

* [overtchat mobile-v1.13.0](https://github.com/yoloyash/overtchat) – Lightweight self-hosted chat client focused on fast startup, low resource use, and BYO API keys or local models, with optional web search and mobile support.

* [Bifrost core/v1.11.3](https://github.com/maximhq/bifrost) – High-performance AI gateway offering multi-provider access with automatic failover, load balancing, and zero-downtime deployments.

* [RunAnywhere cpp-desktop-v0.20.37...](https://github.com/RunanywhereAI/runanywhere-sdks) – On-device mobile SDKs for running LLMs, speech-to-text, and text-to-speech locally.

* [OpenScience v2.0.151](https://github.com/synthetic-sciences/openscience) – Model-agnostic AI workbench that loops through literature review, hypothesis building, code and experiment execution, and research write-ups in a browser workspace.

* [RepoWiki v0.5.1](https://github.com/he-yufeng/RepoWiki) – RepoWiki generates structured wiki documentation for local or GitHub codebases from a CLI or web interface, exporting Markdown, JSON, or self-contained HTML with terminal Q&A.

* [QVAC sdk-v0.21.0](https://github.com/tetherto/qvac) – Local-first, cross-platform SDK for building peer-to-peer AI apps with local model inference, speech, translation, and RAG.

* [Rome app-runtime-v0.7.2](https://github.com/rome-os/rome) – Agentic OS that provides a guardrailed environment for collaborative human and agent workflows with persistent tools, memory, and interfaces.

* [magpie v0.1.1165](https://github.com/yetone/magpie) – Desktop and terminal menu-bar app that lets you view and switch each installed AI agent’s configured model via a local model gateway.

* [GoAI v0.10.8](https://github.com/zendev-sh/goai) – Go SDK providing a unified API for 25+ AI providers with streaming, structured output, and MCP support.

* [AI SDK @ai-sdk/vue@3.0.303](https://github.com/vercel/ai) – TypeScript toolkit for building AI-powered applications with popular frameworks.

* [Casbin Gateway nightly](https://github.com/apache/casbin-gateway) – Casbin Gateway provides an endpoint that tracks AI coding agents’ requests and grades whether the accessed model API matches the paid provider, with per-agent permissions and records.

* [San v1.24.1](https://github.com/genai-io/san) – Terminal-native unified runtime for specialized AI agents built on pluggable LLMs, search backends, personas, and skills.

* [DIAL Chat 0.49.7](https://github.com/epam/ai-dial-chat) – Default UI for DIAL offering conversation interface, IDP support, model side-by-side comparison, extensions and theming.

* [Blade Code v0.11.12](https://github.com/echoVic/blade-code) – AI CLI and web UI coding agent with 20+ tools, MCP support, and configurable providers for model-driven coding workflows.

* [Wegent wework-v0.5.6](https://github.com/wecode-ai/Wegent) – Self-hostable platform for building and running AI agent teams across chat, coding, knowledge, and automation with shared capabilities and local execution via Wework.

* [Meuxe v0.1.6](https://github.com/meet447/Meuxe) – Desktop companion app with local-first memory, expressive Live2D/VRM avatars, and voice chat driven by an installed ACP agent.

* [MoShen v0.11.4](https://github.com/M1r0er/moshen) – A personal writing agent that organizes notes, self-checks settings, and tracks plot foreshadowing without writing for you.

* [OpenEverest v2.0.0-dev.4](https://github.com/openeverest/openeverest) – Cloud-native, multi-database platform for automated provisioning, deployment, scaling, and management on Kubernetes.

* [Headroom v0.9.36](https://github.com/gglucass/headroom-desktop) – Local-first desktop tray app that routes coding clients through a token-saving optimization pipeline and shows savings analytics, primarily for Claude Code.

* [HelixML 2.12.33](https://github.com/helixml/helix) – Private GenAI stack for deploying AI agents with support for RAG, API calls, vision, and efficient GPU scheduling.

* [Full-Stack AI Agent Template 0.2.21](https://github.com/vstorm-co/full-stack-ai-agent-template) – Production-ready FastAPI+Next.js template for building type-safe AI/LLM agents with streaming, auth, observability, multi-DB support, and 20+ integrations.

* [ElevenLabs Monorepo for NPM Package @elevenlabs/client@1...](https://github.com/elevenlabs/packages) – Monorepo managing multiple npm packages under the @elevenlabs scope.

* [scribe v0.5.6](https://github.com/oliver-kriska/scribe) – Single-binary CLI that continuously extracts and indexes git, coding-agent sessions, and iMessage bookmarks into an LLM-written wiki for semantic search.

* [Huddol v0.0.9](https://github.com/ImFeH2/fora) – Huddol lets you run your own organization of agents.

* [sciClaw v0.3.10](https://github.com/drpedapati/sciclaw) – Paired-scientist AI assistant for reproducible research that executes hypothesis-driven workflows, runs tools, and records auditable evidence.

* [Utopia v0.1.0-rc9](https://github.com/deeplethe/utopia) – Passively evolving enterprise world model that captures knowledge, rules, reasoning, and controlled execution over time.

* [I3K RAG Engine v0.1.48](https://github.com/I3K-IT/RAG-Enterprise) – Self-hosted retrieval-augmented generation engine that keeps all document processing and answering offline on a single binary, with cited sources.

* [Dexto @dexto/tools-schedul...](https://github.com/truffle-ai/dexto) – Universal intelligence layer that orchestrates LLMs, tools, and data to build stateful, multimodal agentic applications.

* [Tingly Box v1.261006.2](https://github.com/tingly-dev/tingly-box) – High-performance desktop LLM proxy that unifies access to many models and providers via a single OpenAI-compatible API.

* [Quivr V2 v2.0.0-alpha.6](https://github.com/The-Vibe-Company/quivr) – Durable content ingestion and hybrid search engine for monitoring topics over time, with plugin-based connectors, enrichments, and business rules.

* [OpenBiliClaw extension-v0.3.227](https://github.com/whiteguo233/OpenBiliClaw) – Local, private, self-improving content discovery agent that learns from interactions to find videos and posts across Bilibili, Xiaohongshu, Douyin, YouTube, and more.

* [AgentField v0.1.142-rc.9](https://github.com/Agent-Field/agentfield) – Control-plane infrastructure that deploys, scales, and secures autonomous AI agents as observable, identity-aware backend services.

* [senpi v2026.10.10](https://github.com/code-yeongyu/senpi) – Opinionated TypeScript monorepo fork of pi-mono that provides a curated coding-agent runtime with builtin extensions and core tweaks.

* [Aivory v2.5.1-beta.14](https://github.com/hjxwz123/Aivory) – Production-ready, self-hosted multi-model AI workspace with interleaved tool execution, RAG knowledge bases, persistent Python sandbox, team workspaces, and full admin backend.

* [llama.rn v0.13.0-rc.7](https://github.com/mybigday/llama.rn) – React Native binding for running LLaMA model inference with multimodal support including vision and audio.

* [Stately Agent @statelyai/agent@2.0...](https://github.com/statelyai/agent) – State-machine-powered LLM agent logic built on XState, where model decisions and tool effects run within explicit, inspectable control flow.

* [Rulora v0.1.0-alpha.4](https://github.com/Buffalo2024/Rulora) – Rulora provides hybrid control components for AI agents, including validated outputs, bounded loops, and collective decision boundaries backed by TypeScript and JSON Schema.