Agent Smith
✨A local-first AI agents toolkit for Node.js and the browser, providing composable Agent, Task, and semantic memory modules, compatible with local inference backends like Ollama/Llama.cpp, emphasizing explicit control and declarative definitions.
Model & Inference Framework大语言模型SDK
AgentHub SDK
✨A bilingual unified LLM SDK supporting seamless switching between GPT, Claude, and Gemini models, offering zero-code multi-model integration, standardized tool calling, fine-grained tracing, and built-in Web UI debugging tools.
Model & Inference Framework大语言模型AI Agents
NullClaw
✨Ultra-lightweight fully autonomous AI agent infrastructure with 678 KB single binary, supporting 22+ AI providers and 17+ communication channels.
Model & Inference FrameworkRAGSDK
Inspect
✨An open-source framework for large language model evaluations from the UK AI Safety Institute, featuring a modular Datasets/Solvers/Scorers architecture, multi-model/tool support, sandboxed execution, and 100+ pre-built benchmarks.
Model & Inference FrameworkSDKAI Agents
Edge-Veda
✨On-device full-stack AI SDK for Flutter with LLM, Vision, Speech, Image Gen, and RAG; features compute budget contracts and adaptive QoS with zero cloud dependency.
Rhesis
✨Open-source platform for testing LLM and agentic apps with AI-powered test generation, adversarial red-teaming, and 60+ evaluation metrics for RAG hallucination detection and conversation consistency verification.
Model & Inference Framework大语言模型SDK
Semantica
✨A framework for building semantic layers, context graphs, and decision intelligence systems with explainability and provenance.
Hindsight
✨An agent memory system with biomimetic architecture that enables AI to learn from experience via reflection, featuring multi-strategy retrieval and top performance on LongMemEval benchmark, released under MIT license.
Teleton Agent
✨An autonomous AI agent platform operating as a real Telegram user with native TON blockchain integration (wallet, DEX, NFT, DNS), featuring 100+ built-in tools, multi-provider LLM support, Hybrid RAG memory system, Plugin SDK, and MCP client extensibility.
Model & Inference Framework大语言模型SDK
any-llm
✨A unified LLM access SDK by Mozilla AI connecting to 30+ cloud and local models via a single interface. Supports OpenAI format I/O without needing a proxy server.
Model & Inference FrameworkAI Agents大语言模型