model-agnostic 开源项目推荐（175 个）

aipoch /open-science

面向科学研究与智能体工作流的开源 AI 研究工作台。本地优先、模型无关的桌面应用，具备可扩展技能、MCP 工具与连接器、Python/R 执行以及可追溯产物，支持在 macOS、Windows 和 Linux 上进行可复现研究。

huawei-csl /SINQ

Welcome to the official repository of SINQ! A novel, fast and high-quality quantization method designed to make any Large Language Model smaller while preserving accuracy [ICML 2026]

0xSero /parchi

Your AI friend right in your browser

AmrDab /clawdcursor

clawdcursor将屏幕上的任何内容编译为一个UI地图-将辅助功能树和OCR融合为稳定的、可访问的元素，仅在需要时才提供屏幕截图-然后通过可重复使用的脚本驱动应用程序，验证每个操作并将其路由通过单个安全门。

ntorga /agent-starter-kit

The scaffold for your multi-model, personalized Natural Language AI Harness (NLAH) .

OpenCoven /coven

Local-first runtime for project-scoped AI coding-agent sessions, with durable state, authority boundaries, and multi-harness interoperability.

jdpolasky /chief-of-staff-2

An AI chief of staff in plain markdown, in an Obsidian vault. It tracks your calendar, commitments, and rules, and hands you a morning brief. Version two: torn down and rebuilt smaller. Built on Claude Code, ADHD-first, open to anyone.

reds-lab /LAVA

This is an official repository for "LAVA: Data Valuation without Pre-Specified Learning Algorithms" (ICLR2023).

avasol /galadriel-public

The open engine behind Aedelgard: a persistent, self-hosted AI agent with a verbatim memory palace and a model-agnostic brain. Swap Claude, Gemini, OpenAI, Mistral, DeepSeek or a local model without losing the mind.

iikarus /Dragon-Brain

Dragon Brain — persistent long-term memory for AI agents via MCP (Model Context Protocol). Knowledge graph (FalkorDB) + vector search (Qdrant) + CUDA GPU embeddings. Works with Claude, Gemini CLI, Cursor, Windsurf, VS Code Copilot. 30 tools, 1121 tests.

woai3c /x-code-cli

Open-source, model-agnostic coding agent CLI with subagents, skills, MCP, and plugins.

MarcelRobeer /ContrastiveExplanation

Contrastive Explanation (Foil Trees), developed at TNO/Utrecht University

DenAB-NVS /submarine

Sovereign, depth-adaptive AI memory. Open source. Local-first. Model-agnostic.

AbyssCN /oh-my-dag

The orchestration layer under your coding agent. Turns work into a typed graph, runs one model per node, and takes the verdict from outside the model — exit codes, write-set checks, and a cross-family verifier. MCP server, 50 tools, bring your own models.

quantamixsol /graqle

Persistent organisational intelligence for AI agents. Turn codebases, documents, policies and decisions into a knowledge graph — 13 LLM backends, impact analysis, evidence + confidence. Works with Claude Code, Cursor, Copilot.

diggerhq /opendots

An always-on personal agent you own, on the model you choose. A model-agnostic alternative to OpenAI's Dots, built on OpenComputer Serverless Agents.

hogan-tech /brand-loom

Open-source marketing skills that run on any model — with a hosted brand-memory engine. No coding agent required.

lunarstack-co /agentling

The agent loop for Python. Your harness, any model.

ntakafumi /aime

Approximate Inverse Model Explanations (AIME): model-agnostic XAI through an approximate inverse operator.

FlorianBruniaux /flow-lean

Model-agnostic output control for LLM agents: concise, action-first answers with proof and risk preserved. Native install paths for Claude Code and Codex.