Agent frameworks & infrastructure
LLM gateways & API proxies
4,018 agents · most starred first
LLM gateways & API proxies, page 14 of 41
CLI Proxy API (CPA) Plugin - Prioritize Claude seats reaching their weekly limits
DSH 模型整理(模型列表整理):隐藏用不到的模型、给模型和渠道改备注名、用箭头调整排序,保留厂商计费倍率。Model list tidy for DeepSeek Harness.
一站式纳管 WorkBuddy/Trae/Qoder/千问办公/Qwen/豆包 六大国产 AI 平台:OAuth/扫码/PAT 多方式接入、自动签到攒积分、额度轮询、OpenAI/Responses/Claude 三协议兼容网关、发布模型多渠道路由与故障转移,单二进制零依赖部署
Cost-aware model routing for LangChain agents. Routes execution turns to a fast model, keeps the expensive model for planning and recovery.
llm_adaptive_router: Given the following query and candidate models, selects the most appropriate model
Open-source toolkit for working with LLM APIs. invalid token opencode Provides a unified interface for integrating, testing, and automating workflows with models like Claude, GPT, and others. Supports streaming, async execution, and multiple providers. Built for developers who want flexible, scalable, and vendor-independent AI integrations.
An OpenAI/Gemini/Anthropic API helper in C
A reverse proxy for pools of local/remote AI models.
Unified AI API proxy server - GitHub Copilot, OpenCode Zen, Google Antigravity, AWS Kiro, Factory AI Droids
This is a mock service for multiple AI services.
Self-hosted, OpenAI-compatible LLM gateway in a single Go binary. Routes to OpenAI, Anthropic, Gemini, and local Ollama with failover, two-tier caching (exact + semantic), per-key rate limits, and per-customer spend caps.
Governed Model Gateway & Agentic Recording/Debugging Platform
轻量级 OpenAI→Anthropic 协议转换网关,让多上游接口一键变身 Claude 可用入口。 A lightweight OpenAI-to-Anthropic gateway that turns multiple upstream APIs into one Claude-ready endpoint.
Spora is the lightest OSS AI Gateway. This product is managed by the startup Fisarum.
A single-page Next.js application that validates LLM API keys for multiple providers.
Local Claude Code is a dead-simple installer that rewires Claude Code to talk to any LLM server, like your local LM Studio instance, an Ollama endpoint, a self-hosted vLLM cluster, or any OpenAI-compatible API
Cut Claude Code spend without sacrificing quality — and prove it. Haiku/Sonnet/Opus router with real $-saved numbers, not vibes.
如何将动态变化、经常被训练或计算任务占用的本地消费级 GPU 资源,封装成一个可弹性升降档、持续可用、并对外保持 OpenAI / Anthropic API 接口稳定的大模型推理服务。
Claude中转站推荐,国内直连可试用稳定可靠的Claude API中转站
OpenAI-compatible proxy for Claude Pro/Max - use with Cline, Roo Code, Kilo Code
Autohand Routes - the model router built for coding agents: policy-driven, provider-agnostic, and battle-tested aross millions of sessions.
Complete guide to run Claude Code CLI & VS Code with AgentRouter using a WAF bypass proxy
Cut your Claude API bill: local proxy that auto-routes every request to right model. Zero code changes.
High-performance Rust proxy that translates Anthropic Messages API to OpenAI-compatible format, routing to NVIDIA NIM, Ollama & OpenRouter. Built with Tokio + Axum.
A unified CLI for multiple AI language models with AWS Bedrock integration, MCP support, and autonomous agent capabilities. Access 13+ AI providers (Claude, OpenAI, Gemini, Bedrock, and more) through one interface with secure file operations, approval workflows, and agent-driven automation.
A local-first, single-binary LLM router & Agent flight recorder. Automatically turns byte-faithful logs into structured Agent task stories, compares agent behaviors, and hides provider failover behind virtual endpoints.
LLM API proxy with 3-format conversion (OpenAI Chat/Responses/Anthropic Messages), smart routing, failover & web dashboard. 智能路由,三格式互转,多供应商管理
Local OAuth proxy that auto-rotates Claude Code accounts across usage limits on macOS and Linux.
Unofficial Chrome extension: switch between Claude.ai and Claude Platform accounts, track usage limits, auto-switch by rules. 100% local.
The ultimate AI infrastructure platform: One API, every leading model, zero integration friction.
Open-source library built for fast and reliable connections to different LLM providers
Intelligent middleware platform for Claude Code — route to any model, track costs, cache responses, auto-failover, with a 31-agent AI development team built in
Access premium GPT services on a budget.
Official package for PurGPT.
The official Python library for the ShuttleAI API
GEMINI-SERVER is a tool to access Gemini AI locally on localhost. Users can generate text-based content by providing prompts, without relying on third-party services, using an API Key from Google AI for direct interaction.
Sample demonstrating how to load balance requests between multiple Azure OpenAI Services using Azure API Management.
Gateway to control LLM API/SDK calls. Supports access to OpenAI, Azure, Anthropic and Bedrock calls
Swift Package Manager for accessing the OpenAI GPT-3 API
The gem that makes it easier to use the OpenAI API
Production AI orchestration routing requests between local models (Ollama) and cloud APIs (Claude/OpenAI). Reduces AI costs 60-80% by running lightweight models locally for simple tasks.
A Proxy for Github Copilot that converts OpenAI-compatible code to the Ollama API.
A fast Rust-based gateway unifying access to multiple LLM providers with logging, user control, and key-based authentication.
Enables the application to transition seamlessly between different OpenAI models, choosing the one that is most suitable for the task described in the user's prompt.
A simple npm package to perform requests as a user on the OpenAI ChatGPT page.
PHPOpenAI is a wrapper to access OpenAI API
A lightweight local AI gateway that securely proxies LLM API requests with automatic key injection and CORS support.
It is a project designed as a combination of different AI APIs to make it easy for the user to interact with the capabilities of different AI models. The project contains best practices, is easily maintainable and extensible. The code is built and organized intelligently, divided into its own areas of responsibility.
Easy access to all the AI API's in one place.
GPTAggregator is a Python-based application that provides a unified interface to interact with various large language models (LLMs) through their respective APIs. The project aims to simplify the process of working with different LLM providers.
Windows desktop app to improve LLM requests that use the OpenAI API format
Python script to list all the available OpenAI models in a subscription and show the available quota
Nyquest — Semantic Compression Proxy for LLMs. 350+ rules, local LLM stage, 15-75% token savings. Full Rust stack.
Authenticating with Entra ID (former Azure AD) to access Azure OpenAI models in Python SDK v1.x
【一个类封装】大模型API非流式调用、流式调用、单轮对话、多轮对话(有记忆)代码模板
一个基于 FastAPI 的轻量代理/适配层:将 Anthropic/Claude 的消息与工具调用请求转换为 OpenAI Chat Completions 兼容格式;智能选择工具定义处理策略(系统提示词 vs 用户消息),根据配置自动优化性能和功能完整性;支持可选的自动工具选择、SSE 流式转发,以及将目标模型响应回转为 Anthropic 格式。仅提供服务端代理,不侵入客户端 SDK。
Enable developers to integrate AI chat/copilot capabilities into web apps while ensuring users keep control over which provider/model is used and how data flows.
OpenAI Stream HTTP Adapter - A Go-based server providing OpenAI-compatible API endpoints with full HTTP streaming support. Bridges AI platforms with streaming services, converting SSE to OpenAI format. Perfect for Dify, LangChain, and intelligent agent integrations.
Local-first dashboard for managing multiple OpenAI accounts — track subscriptions, live usage quotas, and agent assignments.
Integrate AIProxy with Expo apps
Open-source AI Gateway, all enterprise features included
Forward any request to the API endpoint of OpenAI.
An OpenAI API proxy running with Cloudflare worker.
An api for listing the x.com api, listing the grok endpoints, and wrapping them into a compatible openai protocol
Function-calling API for LLM from multiple providers
A unified .NET abstraction layer for multiple Large Language Model (LLM) providers. LLM.Nexus simplifies integration with OpenAI, Anthropic, Google, and other LLM services through a common interface, making it easy to switch between providers or support multiple providers in your application simultaneously.
A powerful .NET Nuget Package for interacting with the Herc.ai API.
unified Rust interface for multiple LLM providers
OpenAI-compatible API proxy for Claude Max subscriptions - direct API, fast, reliable
A comprehensive Go library for managing multiple LLM (Large Language Model) providers with intelligent API key management, rotation strategies, and unified interfaces. GoLLMKit simplifies working with OpenAI, Anthropic, Google Gemini, and other LLM providers while providing enterprise-grade features like cost optimization, health monitoring etc
Fusion 后端:统一多模型接入,提供 Agent 对话、文件解析、向量检索与知识库能力。
Professional multi-AI development environment with intelligent cost optimization (<5/month)
An agent-friendly LLM proxy that cleans up Kimi K2 output, fixes tool calls, and normalizes responses for OpenAI/Anthropic-style clients.
ModelMux is a high-performance Rust proxy server that seamlessly converts OpenAI-compatible API requests to Vertex AI (Anthropic Claude) format. Built with Rust Edition 2024 for maximum performance and type safety.
Chat participant is a Visual Studio Code extension that allows you start a little development-grade server on your machine(localhost) to proxy your OpenAI API compliant requests to models behind GitHub Copilot
This python script gives you a fluent output.
Secure API key manager for macOS — Keychain + local proxy keeps keys out of env
Distributed rate limiter for LLM APIs with FIFO queues, TPM/RPM limits, and burst prevention
Stop juggling six AI CLIs. Conclave is your universal remote for LLMs - query any model with one syntax, or unleash them all in parallel and let a judge synthesize the chaos into a verdict. Why trust one robot when you can have a whole council argue about your code?
Lightweight local LLM proxy with SQLite cost tracking & interactive model comparison. Unified OpenAI-compatible API for multiple providers.
Claude API 智能转发代理桌面应用 - 多端点负载均衡、自动故障转移、实时使用统计
Multi-provider LLM chat service — FastAPI + Next.js. Automatic failover (OpenAI → Groq), SSE streaming, response caching with optional Redis, per-request telemetry (provider, cache hit, latency, tokens, cost), demo guardrails. Runs free-tier on Render + Vercel. ADRs and honest benchmarks included.
ZIM-based retrieval augmented proxy for OpenAI-compatible AI
Simple project to connect another project work as either AI connector or AI runtime with openAI API style, great for learning, testing, experiment, without expensive API that work all locally.
A Bun + FastAPI monorepo for running a mobile-friendly AI workspace in the browser, backed by a configurable LLM gateway and repository indexing APIs.
Achieve Frontier AI performance in your CLI — fuse the local model CLIs you already run. Fan a prompt across a panel, a judge model compares the answers, a synthesizer writes one grounded reply instead of a majority vote. On a 100-task benchmark, every fusion panel beat its solo members.
AI API key & subscription manager for Claude Code — your keys never touch the chat
The successor to ds2api
【API中转站推荐】国内直连ChatGPT/Claude API/Claude Code服务_神马中转API最全最便宜的大模型中转API,2025最新稳定好用的API中转平台推荐_AI中转站推荐
Connects WordPress 7.0's AI Client to any OpenAI-compatible endpoint (Ollama, LM Studio, OpenRouter, etc.)
A self-hosted LLM proxy service built around LiteLLM, with support for additional proxy tools and provider integrations such as CLIProxyAPI. Includes monitoring capabilities and is deployed on Docker Swarm, managed via Portainer.
Prism is a modern multi-model AI gateway and subscription platform with an OpenAI-compatible API, model marketplace, tool use, quota windows, billing, admin console, file parsing, web search, and PWA/Tauri clients.
Point your existing SDK at one URL and reach every LLM vendor — with real failover, not a try/except. One static Rust binary.
Free OpenAI-compatible API that forwards requests to GPT via browser extension - for testing
Integrate AIProxy with Expo apps
Local-first API router for MiniMax + Kiro (AWS CodeWhisperer / Amazon Q) upstreams — OpenAI & Anthropic compatible, multi-account fallback, switchable Kiro IDE/CLI persona, RTK + Caveman compression, built-in dashboard (Hono + SQLite + Preact)
Claude Code profile manager and local compatibility proxies. Easily switch profiles and run Claude Code using Gemini, Groq, Mistral, NVIDIA NIM, Hugging Face, OpenRouter, and OpenAI-compatible APIs.
A CLI tool for querying OpenAI-compatible endpoints with a prompt and input file
Xcode Intelligence 使用 deepseek