155 个 · 框架 · AI 核心
基于 Ray 的易用、可扩展、高性能 Agentic RL 框架（PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL）An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)
Cybersecurity AI (CAI)，面向 AI 安全的框架。Cybersecurity AI (CAI), the framework for AI Security
⚙️🦀 使用 Rust 构建模块化、可扩展的 LLM 应用。⚙️🦀 Build modular and scalable LLM Applications in Rust
Evidently 是一个开源的 ML 和 LLM 可观测性框架，评估、测试和监控任何 AI 驱动的系统或数据 pipeline，覆盖从表格数据到 Gen AI 场景，提供 100+ 指标。Evidently is an open-source ML and LLM observability framework. Evaluate, test, and monitor any AI-powered system or data pipeline. From tabular data to Gen AI. 100+ metrics.
FinRobot：基于 LLM 的开源金融分析 AI Agent 平台 🚀 🚀 🚀FinRobot: An Open-Source AI Agent Platform for Financial Applications using LLMs 🚀 🚀 🚀
灵活而强大的框架，用于管理多个 AI agent 并处理复杂对话。Flexible and powerful framework for managing multiple AI agents and handling complex conversations
数分钟内（而非数月）将 AI Agents 部署到 Google Cloud。提供生产就绪模板，内置 CI/CD、评估和可观测性。Ship AI Agents to Google Cloud in minutes, not months. Production-ready templates with built-in CI/CD, evaluation, and observability.
用于构建 agentic apps 的开源框架，支持 JavaScript、Go、Dart 与 Python，由 Google 在生产环境中构建并使用。Open-source framework for building agentic apps in JavaScript, Go, Dart, and Python, built and used in production by Google
Mooncake 是 Moonshot AI 旗下领先 LLM 服务 Kimi 的服务平台。Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
🧊 开源 LLM 可观测性平台，一行代码即可实现监控、评估与实验。YC W23 🍓🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓
AI 原生网络安全的执行系统——意图转化为受治理的执行，证据成为可运营的记忆，每一次操作都让下一次更优。The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.
下一代 AI Agent 优化平台：Cozeloop 提供从开发、调试、评估到监控的全生命周期管理能力，解决 AI agent 开发中的挑战。Next-generation AI Agent Optimization Platform: Cozeloop addresses challenges in AI agent development by providing full-lifecycle management capabilities from development, debugging, and evaluation to monitoring.
用于构建复杂创新 RAG 流水线的低代码 MCP 框架A Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines
ZenML 🙏：从 Pipeline 到 Agent 的一站式 AI 平台。https://zenml.ioZenML 🙏: One AI Platform from Pipelines to Agents. https://zenml.io.
cube studio开源云原生一站式机器学习/深度学习/大模型AI平台，mlops算法链路全流程，算力租赁平台，notebook在线开发，拖拉拽任务流pipeline编排，多机多卡分布式训练，超参搜索，推理服务VGPU虚拟化，边缘计算，标注平台自动化标注，deepseek等大模型sft微调/奖励模型/强化学习训练，vllm/ollama/mindie大模型多机推理，私有知识库，AI模型市场，支持国产cpu/gpu/npu 昇腾生态，支持RDMA，支持pytorch/tf/mxnet/deepspeed/paddle/colossalai/horovod/ray/volcano等分布式
开源 LLMOps 平台：集成 prompt playground、prompt 管理、LLM 评估和 LLM 可观测性。The open-source LLMOps platform: prompt playground, prompt management, LLM evaluation, and LLM observability all in one place.
针对加速基础设施和微服务架构优化的生成式 AI 参考工作流。Generative AI reference workflows optimized for accelerated infrastructure and microservice architecture.
任意模型，任意硬件，零妥协。基于 @ziglang / @openxla / MLIR / @bazelbuild 构建。Any model. Any hardware. Zero compromise. Built with @ziglang / @openxla / MLIR / @bazelbuild
用于加速视频生成的统一推理与后训练框架。A unified inference and post-training framework for accelerated video generation.
面向 LLM 评估和 AI agent 测试的平台The platform for LLM evaluations and AI agent testing
Apple Silicon 上最快的本地 AI 引擎。比 Ollama 快 4.2 倍，缓存 TTFT 仅 0.08s，工具调用支持率 100%。内置 17 种工具解析器、prompt cache、推理分离、云端路由。可作为 OpenAI 的即插即用替代，兼容 Claude Code、Cursor、Aider。The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separation, cloud routing. Drop-in OpenAI replacement. Works with Claude Code, Cursor, Aider.
AGiXT 是一个动态 AI Agent 自动化平台，无缝编排跨多个 AI provider 的指令管理和复杂任务执行，结合自适应记忆、智能特性与可扩展插件系统，提供高效全面的 AI 解决方案。AGiXT is a dynamic AI Agent Automation Platform that seamlessly orchestrates instruction management and complex task execution across diverse AI providers. Combining adaptive memory, smart features, and a versatile plugin system, AGiXT delivers efficient and comprehensive AI solutions.
Laminar —— 专为 AI agent 打造的开源可观测性平台，YC S24。Laminar - open-source observability platform purpose-built for AI agents. YC S24.
几乎可视为 DSPy 在 TypeScript 上的"官方"框架。The pretty much "official" DSPy framework for Typescript
开源 AI 工程平台：基于 OpenTelemetry 的 LLM 可观测性、GPU 监控、guardrails、评估、prompt 管理、Vault、Playground。🚀💻 集成 50+ LLM provider、VectorDB、agent 框架与 GPU。Open source platform for AI Engineering: OpenTelemetry-native LLM Observability, GPU Monitoring, Guardrails, Evaluations, Prompt Management, Vault, Playground. 🚀💻 Integrates with 50+ LLM Providers, VectorDBs, Agent Frameworks and GPUs.
Apache Hamilton 帮助数据科学家和工程师定义可测试、模块化、自文档化的数据流，内置 lineage/tracing 与 metadata，可在任何支持 Python 的环境中运行和扩展。Apache Hamilton helps data scientists and engineers define testable, modular, self-documenting dataflows, that encode lineage/tracing and metadata. Runs and scales everywhere python does.
cubestudio开源云原生一站式机器学习/深度学习/大模型AI平台/MaaS/mlops/人工智能平台/训推平台，算法全链路流程，多租户，算力租赁平台，token中转，拖拉拽任务流pipeline编排，多机多卡分布式训练，超参搜索，推理服务，VGPU虚拟化，云边端协同，边缘计算，自动化标注平台，deepseek等大模型sft微调/奖励模型/强化学习训练，vllm/ollama/mindie大模型多机推理，私有知识库llmops智能体，AI模型市场，支持国产异构算力调度,昇腾/寒武纪/海光/摩尔/沐曦等，支持ib/roce/RDMA，信创支持
一个简洁、高性能且可扩展的 Jax LLM！A simple, performant, and scalable Jax LLM!
基于 Envoy Gateway 构建，提供生成式 AI 服务的统一接入管理Manages Unified Access to Generative AI Services built on Envoy Gateway
开源、端到端的 LLM 与 AI agent 应用评估、观测与改进平台。Tracing · Evals · Simulations · Datasets · Gateway · Guardrails。可自托管，Apache 2.0 许可。Open-source, end-to-end platform for evaluating, observing, and improving LLM and AI agent applications. Tracing · Evals · Simulations · Datasets · Gateway · Guardrails. Self-hostable. Apache 2.0.
通过统一 API 接口路由、管理和分析跨多家服务商的 LLM 请求Route, manage, and analyze your LLM requests across multiple providers with a unified API interface.
一个使用模拟的、真实的合成交互对 Agent 进行全面诊断和优化的框架。A framework for comprehensive diagnosis and optimization of agents using simulated, realistic synthetic interactions
面向 LLM 和 LVLM 的强大模型压缩框架，适配 NVIDIA GPU 和华为昇腾 NPUA powerful model compression framework for LLMs and LVLMs, adapted for NVIDIA GPUs and Huawei Ascend NPUs.