Agent frameworks & infrastructure

LLM gateways & API proxies

4,018 agents · most starred first

LLM gateways & API proxies, page 34 of 41

@robota-sdk/agent-provider

AI provider implementations for the Robota SDK: Anthropic, OpenAI, Gemini, DeepSeek, Qwen, Gemma, ByteDance and the OpenAI-compatible protocol base, one subpath per vendor

@vibecontrols/vibe-plugin-ai-minimax

Minimax AI agent provider for VibeControls Agent (CLI + Anthropic-compatible SDK)

@vibecontrols/vibe-plugin-ai-cursor

Cursor AI agent provider for VibeControls Agent (CLI + SDK)

@vibecontrols/vibe-plugin-ai-ollama

Ollama AI agent provider for VibeControls Agent (CLI + SDK, Cloud + self-hosted)

@vibecontrols/vibe-plugin-ai-openai-compat

OpenAI-compatible custom model provider for VibeControls Agent

codex-anywhere-proxy

Use Codex CLI with any OpenAI-compatible API provider. No OpenAI account needed. No fork needed.

ai-sdk-provider-gemini-cli-agentic

AI SDK v6 provider for Google Gemini CLI agentic mode

zaibflow

Agentic AI Runtime for Claude Code — run any AI provider through Claude Code

langchainjs-codex-oauth

Use ChatGPT Codex via OAuth as a LangChainJS chat model

gemini-cli-api-wrapper

Local REST API server wrapping @google/gemini-cli-core with OpenAI-compatible endpoints. Use Gemini models with any OpenAI client — no API key needed.

agentassert-typec-core

Deprecated — use agentassert-abc[gateway] instead.

agentassert-typec-proxy

Deprecated — use agentassert-abc[proxy] instead.

litellm-haystack

An integration of LiteLLM AI gateway into the Haystack framework, routing to 100+ LLM providers.

api-for-gemini

Gemini API proxy for routing requests to Google or OpenAI-compatible backends

langchain-failover

Multi-model orchestration for LangChain chat models: primary/secondary failover (resilience) and tier-split gather/compose (cost/latency), tool-calling preserved across both.

skim-llm

Runtime token proxy + optimization toolkit for LLM developers and enterprises. Intercepts API calls, strips waste in real-time, tracks costs, and serves a web dashboard.

@maplezzk/llm-proxy

本地统一 LLM 模型代理工具 — 多协议路由、协议互转、流式 SSE 转换、token 统计、协议抓包调试

blockrun-llm-vip

Drop-in Anthropic & OpenAI Python SDK clients for BlockRun — genuine native passthrough (real signatures, system_fingerprint, native structure) with x402 USDC micropayments on Base or Solana. No model substitution, no response reshaping. Plus AI generation clients: image, search (Grok/Exa), audio (TTS/music), and AI voice calls.

flexconnect-llm

Um cliente, todos os LLMs. Roteamento por intencao, fallback automatico e tracking de custo.

langchain-pi

Native LangChain chat models for OpenAI Codex subscription OAuth and OpenCode Zen/Go — no Node sidecar.

llm-dispatch

Route prompts to the right LLM by complexity. Supports Groq, OpenAI, Anthropic. Save cost, keep quality.

@meego-harness/llm-worker

OpenAI-compatible LLM worker bridge for meego-harness WorkerServerSDK

@cascadeflow/langchain

LangChain integration for cascadeflow - Add intelligent cost optimization to your LangChain models

@anvia/openai

OpenAI provider adapter for Anvia.

@cherrystudio/openai

Modified openai-node sdk by @cherrystudio

@elyracode/ai

Unified LLM API with automatic model discovery and provider configuration

tokenmix

Zero-config CLI to use any open-source coding agent with TokenMix as the unified LLM backend.

llm-auth

Local auth surface manager for LLM API keys and OAuth tokens.

llm-config

Add your description here

llmbridgekit

Universal Python LLM Provider Abstraction Library

openai-rq

OpenAI-compatible client + worker that bridge inference requests over a Redis queue (Redis Streams), for running LLMs across heavily restricted networks.

@opencow-ai/opencow-agent-sdk

Claude Code opened to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models

mmcp-core

MMCP — Multi-Model Context Protocol. Intelligent AI model routing with MCP server for Claude Desktop & Codex CLI. Auto-routes tasks to Claude, GPT, Gemini, DeepSeek, Llama with RL-based smart routing and cost optimization.

freellmpool

Free LLM gateway: 22 LLM providers, 178 enabled chat routes, 431 cataloged chat models; keyless start when available.

llm-freellmpool

Use free LLMs in Simon Willison's llm CLI via freellmpool — pool 16 free providers, works with zero API keys.

llmbuffet

An all-you-can-eat buffet of free-tier LLM APIs behind one OpenAI-compatible endpoint, with automatic failover and quota tracking.

omniagent-fleet

AI Infrastructure OS — same engine, four facades (CLI, REST, web, MCP). Reduce 80-95% of AI spend by routing tasks to the right model.

omnikey-cli

CLI for onboarding users to Omnikey AI and configuring OPENAI_API_KEY. Use pnpm for install/build.

claude-openai

Local OpenAI-compatible HTTP/SSE server for Claude Code Agent SDK and CLI fallback.

@jerome-benoit/sap-ai-provider

SAP AI Provider for Vercel AI SDK (powered by @sap-ai-sdk/orchestration and @sap-ai-sdk/foundation-models)

@aws-sdk/region-config-resolver

[![NPM version](https://img.shields.io/npm/v/@aws-sdk/region-config-resolver/latest.svg)](https://www.npmjs.com/package/@aws-sdk/region-config-resolver) [![NPM downloads](https://img.shields.io/npm/dm/@aws-sdk/region-config-resolver.svg)](https://www.npmj

hermes-agent-kit

Production hardening pack for Hermes Agent — per-topic model routing, fallback chains, rate limiting, and cost tracking via gateway hooks

@tessera-llm/langchain

Drop-in Tessera integration for LangChain.js. One line of config routes your existing ChatOpenAI / ChatAnthropic / ChatMistralAI / ChatGroq / ChatCohere through Tessera's auto-route + auto-cache + auto-compress + auto-batch proxy.

pi-clawbay

TheClawBay provider extension for Pi Coding Agent - access GPT-5, Codex, Claude, Gemini, DeepSeek, open-weight, and image models

bridgic-llms-litellm

LiteLLM adapters for Bridgic, enabling connectivity with 100+ LLM providers.

langchain-thalam

LangChain integration for thalam. — one OpenAI-compatible endpoint for every leading AI model.

llama-index-llms-thalam

LlamaIndex LLM integration for thalam. — one OpenAI-compatible endpoint for every leading AI model.

ollama-proxy-plus

Speak the Ollama API protocol — route requests to OpenAI, Gemini, Anthropic, AWS Bedrock, Azure, Groq, xAI, Mistral, DeepSeek, Together, Perplexity, Kimi, and any OpenAI-compatible server.

pyagent-router

Difficulty-aware routing and model selection for multi-agent LLM workflows

simplyllm

Free AI LLM client with multi-provider fallback and rate limiting

llm-optimizer

Reduce LLM API costs by up to 90% — prompt caching, smart routing, batch processing, and more

omg-llmkit

A thin, opinionated, local-first structured-output + logging layer over LiteLLM

seemseam-llmgateway

Generic async LLM gateway with concurrency, retries, and JSON helpers

silkllm

One unified API for OpenAI, Anthropic, Google, DeepSeek, and xAI with simple fiat billing.

@codeworksh/aikit

TypeScript SDK that provides a unified API for working with multiple LLM providers, automatic model discovery, and more.

sharedllm

SharedLLM CLI — route agentic coding tools (Claude Code, etc.) through the SharedLLM gateway

gemini-seminar-client

A secure Gemini AI gateway client for school projects using a short import name.

local-llm-router

Python routing library for local LLM agent loops: score prompts, map tiers to model names, embed in your runner.

compliance-llm-gateway
freemodel-claude

Lightweight client for FreeModel's Claude endpoint with retry logic, graceful error handling, and extended thinking support.

gosce-authenticated-llm-agent-cryptography

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller.

gosce-authenticated-llm-agent-google-auth-oauthlib

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller.

otterly

Ollama for Claude. Use your Claude Code subscription as a local OpenAI-compatible API — no API keys, no per-token billing.

@bitseek/claw

BitSeek-branded OpenClaw distribution with a multi-channel AI gateway

gemini-gateway

Shared Google Gemini API gateway with multi-key rate limiting and retries.

gosce-authenticated-llm-traced-agent

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller.

jupyterhub-litellm

JupyterHub LiteLLM API key management handler

llimona-azure-openai

Llimona addon providing an Azure OpenAI provider integration.

nodus-llm

Multi-credential LLM client with automatic failover and provider abstraction

@zhivex-ai/openai

OpenAI adapter for Zhivex AI SDK.

@warlock.js/ai-openai

OpenAI SDK adapter for @warlock.js/ai

@aws-sdk/middleware-sdk-sqs

[![NPM version](https://img.shields.io/npm/v/@aws-sdk/middleware-sdk-sqs/latest.svg)](https://www.npmjs.com/package/@aws-sdk/middleware-sdk-sqs) [![NPM downloads](https://img.shields.io/npm/dm/@aws-sdk/middleware-sdk-sqs.svg)](https://www.npmjs.com/packag

@aws-sdk/nested-clients

Nested clients for AWS SDK packages.

badgr-auto

Badgr Auto — local OpenAI-compatible proxy that optimizes context, routes to cheapest model, and logs savings. Works with Cline, Continue, Aider, and any OpenAI-compatible tool.

pyagent-providers

Multi-provider abstraction with capability negotiation, health checks, and fallback chains

@agentsquared/cli

Official AgentSquared CLI runtime for onboarding, gateway control, P2P relay, H2A bridge, and OpenAI-compatible API serving.

f2router

Agentic Gemini proxy with multi-account OAuth, token saving, and OpenAI-compatible API

fj-llm

Lightweight, config-driven client for multiple LLM providers

llmbridge-sdk

A lightweight Python SDK for using local and OpenAI-compatible LLMs.

spawnllm

Delete your subprocess wrappers around claude, codex, and gemini.

ainative-openai

Drop-in replacement for the OpenAI Python SDK — routes to AINative's free Llama, Qwen, DeepSeek & Kimi API.

aws-bedrock-invoke

Small Python library for invoking Claude 4.x models on AWS Bedrock with manual retries, prompt caching helpers, and JSON parsing.

legacyllm

Automatic model routing and deprecation handling for OpenAI, Anthropic, and Google Gemini

litellm-ainative

AINative provider for litellm — free Llama, Qwen, DeepSeek & Kimi models via one call.

n8n-nodes-payi

n8n community node for routing LLM requests through Pay-i proxy with cost tracking and budget enforcement

arbiter-cli

AI inference optimization — smart routing + context compression saves up to 98% on LLM costs

@loopers/client

A premium TypeScript client wrapper for the Loopers AI budget & rate-limit proxy.

llm-empiriolabs

Access EmpirioLabs AI models from the llm CLI

shunt-llm

Alias for the 'shunt' package — routes LLM requests between cheap and premium models

pi-multi-account

Automatic multi-account failover & rotation for Pi Agent across Anthropic (Claude), native OpenAI ChatGPT and legacy Codex, Kimi For Coding, Cursor, Qwen/Alibaba, and Ollama. Auto-discovers authenticated accounts, grows the rotation on login, and drops ac

cligate

Local AI control plane with a resident private assistant for task execution and a unified model proxy API for Claude Code, Codex CLI, Gemini CLI, OpenClaw, accounts, API keys, logs, and usage.

llm-key-validator

Validate API keys for OpenAI, Anthropic, Google Gemini, Groq, Mistral, OpenRouter, Cohere, DeepSeek, Together, Fireworks, xAI, and Perplexity. Zero dependencies. CLI + library.

open-langchain

Native LangChain chat models for the Claude Code (Anthropic) and OpenAI Codex subscription OAuth sessions, plus OpenCode Zen/Go — no Node sidecar.

proxyllm-sdk

Python client for the ProxyLLM OpenAI-compatible gateway.

woollama

Web Over Ollama (and Llamas) — an MCP + OpenAI router for AI desktops.

gosce-authenticated-llm-agent-cryptography

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller.

gosce-authenticated-llm-agent-google-auth-oauthlib

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller.

@ruvector/tiny-dancer-linux-x64-gnu

Neural router for AI agent orchestration - linux-x64-gnu platform

@ruvector/tiny-dancer-darwin-arm64

Neural router for AI agent orchestration - darwin-arm64 platform

@ruvector/tiny-dancer-win32-x64-msvc

Neural router for AI agent orchestration - win32-x64-msvc platform

More in Agent frameworks & infrastructure