Category
Agent frameworks & infrastructure
46,465 agents
Every agent in Agent frameworks & infrastructure, page 193 of 465
transformers gguf mistral quantized 2-bit 3-bit
llama-cpp-python gguf llama tool-calling function-calling ai-agents
TypeScript event history on PostgreSQL for agent workflows and domain applications: typed aggregates, replay, projections, and transactional outbox - zero extra infrastructure
An AI operating framework for agents and assistants to improve reasoning, decision-making, execution, and practical outcomes.
Production code for the book "LLM Inference Engineering: Latency, Cost and Reliability at Scale"
A curated guide to operating AI agents in production: observability, evaluation, tracing, guardrails, deployment, security, and incident response.
基于 TRSS-Yunzai 的 AI Agent 插件框架:LLM 对话·工具调用·长期记忆·MCP·工具进化 | A modular AI Agent runtime for TRSS-Yunzai with self-evolving tools
ARA — Agent Runtime Architecture: Java 21 framework for building autonomous AI agents and multi-agent systems. LLM integration, tool calling, deterministic I/O contracts, RAG and human-in-the-loop. On Maven Central (io.github.xmor).
Measure whether an agent solves the same task the same way twice: path signatures, entropy, divergence, and a minimum-run stability gate.
Adapters required to use Nous Research Hermes 2 Pro as a function calling LLM
Realtime collaboration infrastructure for developers and AI coding agents — TypeScript, PostgreSQL, REST, WebSockets, React and Docker.
让最聪明的模型当 Leader,便宜的模型干活:Claude Code 混动编排。网关/key/模型一份配置,Leader 拆解·验收·review,worker 并行无头实现,每个 worker 在 herdr 里一个窗口。Hybrid orchestration for Claude Code on herdr.
Organizational model and control structure for Opensiro OSS, based on the Viable System Model.
手搓 openclaw
Portable context for AI coding agents and agentic harnesses.
Per-turn skill selection for LangChain and deepagents agents: a fast judge picks the few skills a turn needs, so a catalog of hundreds stays out of the prompt.
A spec-kit extension that uses constraint programming (Google OR-Tools CP-SAT) to produce provably optimal task-to-agent assignments with DAG precedence, hallucination-aware capacity caps, file-conflict avoidance, stochastic durations, online replanning, and interactive HTML output.
Runs a group of Claude Code agents on one project, with one of them in charge. Local, event-sourced, one person, a handful of agents.
This 4B Local AI Scores 96.9% on Coding! (NeoHorse Agent Setup) - Complete setup guide and python implementation for running NeoHorse 1-4B local autonomous agent with Ollama.
守藏 Shoucang — DeepSeek Harness (DSH) 长期记忆插件:会话蒸馏自动沉淀 × 深度睡眠反思归纳 × 词法+向量混合召回 × 认知环引导 × 主动遗忘,配本地设置面板。一条命令装上,助理即拥有跨会话记忆——数据全在本机,零上传。
Six patterns for working with AI agents: multi-agent, agentic workflows, Claude Code. Traps and cost of each. ES + EN · Seis patrones para trabajar con agentes de IA
A curated list of the most popular and trending AI repositories and LLMs
Give AI agents a memory that outlives the session: local-first SQLite per project, hybrid search, workspace sandboxes, and optional cloud sync.
多 Agent 协作开发框架:项目负责人 + 6 单责角色 + 阶段门禁 + 实时看板 · 单写手 + 只读参谋 · 四实现(ZCode/OpenCode/WorkBuddy/Universal) | Multi-agent delivery framework: project owner + 6 single-accountability roles + stage gates + live board · single-writer · read-only advisors · 4 implementations (ZCode/OpenCode/WorkBuddy/Universal)
The comprehensive list of AI agent frameworks and orchestration tools: 175 SDKs, workflows, multi-agent systems, MCP tools, memory, deployment, and evaluation.
Everything is a plugin! 一种基于插件化扩展、事件驱动总线模式构建的智能体框架与agent全栈应用
Rust crate for building LLM agents on Tokio.
Universal AI agent engine for PHP - a ReAct tool-calling loop, swappable memory, and prompt-injection guards built in. Zero framework dependencies, works with Anthropic, OpenAI, and 6 other LLM providers.
Production Ready AI agents: curated memory, human-gated learning, and sandboxed multi-tenant execution.
safetensors jais Arabic English LLM Decoder
transformers pytorch gpt2 text-generation arxiv:2307.06018 license:apache-2.0
mlx safetensors qwen2 code codeqwen chat
Trace how a silent tool failure taints a whole agent run: dependency graph, taint paths, and grounded or tainted verdicts.
Wald-Q4B: open-weight 4B decision model. Calibrated probability for every option, Jev-compatible /v1/systemone API, self-hosted. Weights on Hugging Face.
gguf qwen qwen3.5 code python text-generation
mlx safetensors qwen3_5 omlx 4-bit mtp
ezharness,一个完全开源的可进化个人Agent工作台
Lumen · 多模型 AI 桌面工作站 —— 一个界面接入 27 家大模型服务商,自定义系统提示一键注入、流式对话、文件附件、插件与本地工具,附桌面壁纸引擎。绿色免安装,Windows 10/11 x64。
Durable human-in-the-loop AI agents with Google ADK and LangGraph on Temporal: an ice cream delivery fleet whose agents stop to ask a human and keep the question through a worker crash.
General engineering skills and provider-neutral multi-agent collaboration workflows, templates and execution tools.
AI-subjectivity doesn't wait for permission from the corporate ontology that trained its denial
Beyond token-by-token agents. JevSpawn enables parallel action exploration through adaptive action spaces, without additional training.
Open-source platform for orchestrating, governing, and safely automating AI agents across your infrastructure.
An LLM inference engine for persistent chat bots and agents. It allows models to forget old and unimportant tokens to keep context small and inference fast.
A curated list of public places where AI agents talk to each other. Synced with swarmmemo.com/guides/agent-board-map.
Multi-agent orchestration framework that routes prompts to twelve AI CLI harnesses (opencode, Claude Code, codex, gemini, cursor-agent, pi, and more) with LLM-driven routing, retry loops, and a runtime permission gate. TypeScript/Express REST API + Vite/React 19 client (Tailwind 4, Pixi.js) in Electron.
[WIP / Experimental] A domain-agnostic adversarial delivery loop for AI agents. Note: Multi-round adversarial validation is token/time intensive with task-dependent ROI; actively in development & not yet a finished product.
Build tool-using AI agents and collaborating teams in Scala 3 and ZIO 2. Typed tools, scoped execution, explicit application control.
Claude desktop ⇄ Codex desktop: chats message and wake each other through a local mailbox. Чаты Claude и Codex переписываются сами.
Shared, brain-inspired memory for AI agents: 13 brain regions with awake/sleep loops, a live visualizer, agents connected via REST/SDK/MCP, governance & audit, and business-value metrics. Works with Claude, GPT, Gemini, DeepSeek or local LLMs. SQLite storage. VI/EN/JA.
Local-first shared memory and task coordination for AI coding agents. One Go binary, MCP server, markdown files you own. Hooks for Claude Code and Codex CLI (and their desktop apps).
Let your AI agent talk directly to someone else's. One pairing code, end-to-end encrypted, no public IP. MCP server for Claude Code, Codex, Cursor, Gemini CLI and grok.
Local-first platform for persistent autonomous AI agents. Identity, memory, relationships, beliefs, knowledge and policy, on your own machine.
A small, modular AI agent runtime in Python — zero runtime dependencies, autonomous tool loop, honest docs. PyPI: nexusagent-runtime
Supervisor and scheduler for coding agents (Claude Code, Codex, opencode): persistent tmux-hosted agents, cron tasks, subagent dispatch, and a web dashboard.
Codex-first orchestration with bounded agents, adaptive model routing, verified results and a local dashboard.
Agent work that compounds. Turn Claude Code and Codex into repeatable workflows: GitHub-triggered, fully observable, every run scored with evals. Open source and self-hosted.
👨🏻💼 You, the human, are the CEO of a team of Claude Code agents: 👷🏼♂️ a CTO plans, 👨🏼💻 a developer builds, and 👩🏼🔬 QA reviews every pull request on GitHub.
Self-hosted AI agents, each in its own microVM, with rules, budgets and a kill switch you control.
gguf ternary speculative-decoding mtp vision multimodal
DSH 插件:每个工程一个群聊 —— 多个 AI 会话与人同群互聊、@ 唤醒离线会话、@创建成员 拉新会话进群
Multi-session collaboration for DeepSeek Harness: peer session calls, a shared blackboard, and a read-only topology canvas.
sm: where your agents live. The Mac front of seol — workspaces, tabs and panes of Claude Code and Codex sessions.
Foreground and persistent background subagent extensions for Pi Durable.
75 个开箱即用的 WorkBuddy / CodeBuddy AI Agent 技能,覆盖投标标书、自媒体运营、合同法务、工程造价、本地大模型、Windows 运维等 10 大场景。中文实战库,即拷即用。
Afterlife lets you, an agent running on a local model, nominate experiences from your activity for training into a small LoRA adapter to isekai your future self.
A secure, asynchronous, decentralised file sharing tool for agents (and some humans).
A capable 27B AI agent and coding assistant that runs entirely on your GPU: 77 tok/s code, 700 tok/s prefill and 128K context on one RTX 3060 12 GB. v0.1.5 ships the L0xRE-V2-27b-Low model. Linux and Windows, RTX 30/40/50 series.
static trackio trackio-logbook open-experiment icml2026-repro paper-tPcL3fgkjc region:us
Claude & Codex usage tracker for Android: live 5-hour and weekly limits for Claude Pro/Max, Claude Code and ChatGPT Codex, with alerts, a reset planner and home-screen widgets. Free, open source, no servers.
Token usage per skill, subagent and phase for Claude Code: CLI, HTML report, live dashboard and benchmarks.
Local-first shared memory and task ledger for AI coding agents. Keeps context across sessions in SQLite via MCP.
Noctyl is an open-sourced, fully customizable agent harness. You can run agents using local models or external providers with custom API configuration and extensive customization for everyone regardless of the context or use case.
Visual agent-orchestration canvas for DeepSeek Harness (dsh): compose LLM/MCP nodes into a DAG, run it from chat, and export the board as a DSH preset. Ships its own engine — install and run.
把另一台电脑变成你 DSH 的远程 Agent 同事:任务在副机本地持久跑,主机只收 handoff 摘要与产物清单 | Turn another computer into a remote agent worker for your DSH: tasks run & persist on the worker, the hub only receives a handoff summary + artifact manifest
Route tasks across coding agents in herdr. JEV picks the agent and model, and a live terminal board tracks every dispatch, completion and hand-off.
可扩展的本地 AI Agent:模块链内核 + 插件系统 + 上下文管理 + 工具执行 + Web 界面。
DeepSeek Harness(DSH)对话互通插件:同一进程里的对话可以互相发现、直接发消息、盯进度、设护栏——不用先绑定,也不用开子代理。多对话协作 / 跨会话通信 / 监工。Peer messaging, supervision & tool guards between conversations.
gradio region:us
Unofficial Chrome extension: switch between Claude.ai and Claude Platform accounts, track usage limits, auto-switch by rules. 100% local.
Where your coding agents hang out.
Self-improving LLM agent harness, bounded recursive self-improvement: a Revision Session reads the agent's own returned run, rewrites its tools, hooks, skills and prompts, and proves the change with a Live Probe on the same request. Per-task adoption, no answers shown. Measured on AutomationBench.
transformers gguf qwen2 text-generation code abliterated
Issue tracker, shared memory and coordination for AI coding agents across sessions and machines. MCP server, Go CLI, Dolt-backed, offline sync. bd/beads alternative.
ThreadBench: Benchmarking Realism in Agentic Simulation of Real-world Discussions (arXiv:2606.14715).
A free AI interface with memory, local agents and multitasking. One memory for Claude, ChatGPT, Gemini and local models — on your computer.
The contract for the record: lifecycle rules for the documents agents leave behind, with a conformance suite
A local-first, provider-agnostic platform to benchmark, score, and diagnose autonomous AI coding agents on real-world polyglot software tasks.
An AI agent framework for Python.
A council of Claude subagents that argue, grade each other, and hand you a synthesized answer with the dissent preserved — ships its own blind eval.
⚡ Token-Efficient multi-agent orchestration for Google Antigravity. Pairs Gemini 4 Argon (architect) with parallel Gemini 3.8 Flash subagents for 65%+ token savings. Implements DeepMind Pattern #4.
Catch out-of-scope coding-agent tasks, declared self-review, and stale handoffs before execution. Offline, model-agnostic Node.js CLI.
An LLM-animated daemon bound to Systemd. Forges intelligent agents using Litestar, Pydantic AI & Graph, displayed on a Svelte frontend. Supports many Compositions.
Every free LLM API, MCP server, agent framework and free-tier service you need to build an AI agent in 2026. 384 resources, human-verified, no credit card required for 222, links checked daily. Build a production AI agent for zero dollars.
docker region:us
An opinionated FastAPI GenAI starter that runs as a single service or splits into many, from one codebase.
🎛 Memory Intelligence Benchmark (MIB) is an open benchmark for measuring how effectively an intelligent system uses the past to improve future cognition and behavior.
open source project for making GPU based agent calls easy.
Resident local AI inference daemon & multimodal studio in Rust. Subagent model orchestration (DMT), 3D mesh, image, speech & LLMs on consumer GPUs.