mudler/LocalAI
LocalAI is an open-source AI engine that enables running various AI models (LLMs, vision, voice, image, video) on any hardware, including CPU-only systems, with a composable and extensible architec...
Awesome AI Agents › LLM Framework Tools
agentscope-ai/ReMe
ReMe is a local-first long-term memory layer for AI agents, developed in the AgentScope ecosystem and published on PyPI as reme-ai for Python 3.11 and newer. Its central idea is memory as file: memory nodes are plain markdown files with frontmatter and wikilinks that both users and agents can read and edit directly, avoiding an opaque database. On top of that store, the system continuously indexes, links and consolidates memory for later recall. Three automatic processes, described as Auto Memory, Auto Resource and Auto Dream, progressively turn conversations and ingested resources into long-term memories and build wikilink relationships between them without manual curation. Retrieval is progressive and hybrid, combining wikilink traversal, BM25 keyword matching and embeddings for semantic recall, so relationship expansion and lexical matching complement each other; embeddings are optional and disabled by default, and environment variables configure LLM-powered memory evolution and embedding retrieval when wanted. Integration is aimed at being agent-friendly through a SKILL.md file plus a command-line interface, letting different agents read, write, maintain and reuse the same memory store, and a plugin is provided for Claude Code. The documented use cases are giving personal assistants a user-editable long-term memory layer, preserving coding style, project background, repository decisions and workflow experience across sessions for coding agents, turning conversations, notes and resources into a searchable and traceable linked markdown knowledge base, and supporting self-evolving agents that learn from experience by saving successful paths, failed attempts, reusable procedures and periodic reflections. The project descends from earlier MemoryScope and ReMe releases, publishes benchmarks including one on experience-driven agent tool use, and its procedural memory framework was accepted to Findings of ACL 2026. Optional cookbooks cover paper discovery and file-native ETF event research.
https://github.com/agentscope-ai/ReMe
Also in LLM Framework Tools
langfuse/langfuse
Langfuse is an open source LLM engineering platform that enables collaborative development, monitoring, evaluation, and debugging of AI applications with comprehensive observability, prompt management, and integration features.
567-labs/instructor
Instructor is a popular Python library that simplifies working with structured outputs from large language models by providing validation, retries, and streaming support across multiple LLM providers.
TEN-framework/ten-framework
TEN Framework is a real-time, distributed, cloud-edge collaborative multimodal AI agent framework supporting multiple programming languages for building high-performance AI applications.
google/adk-samples
A collection of sample agents built with the Agent Development Kit (ADK) to accelerate development of conversational and multi-agent workflows, designed for customization and integration with Google Cloud services.
truefoundry/cognita
Cognita is an open-source, modular, and production-ready RAG framework by TrueFoundry that simplifies building, customizing, and deploying scalable Retrieval Augmented Generation applications with a no-code UI and extensive integration features.
lemony-ai/cascadeflow
In-process Python and TypeScript library that cascades model calls inside the agent loop, routing each query or tool call between small and flagship models by quality, cost, latency and budget.
shroominic/codeinterpreter-api
An open-source API that implements the ChatGPT Code Interpreter for sandboxed Python code execution, enabling data analysis, visualization, and interactive AI-driven computational tasks with local execution and scalable production deployment.