Agent frameworks & infrastructure
LLM gateways & API proxies
4,018 agents · most starred first
LLM gateways & API proxies, page 41 of 41
Build image, video, and text generation with one AI API. Access leading models through one key, one billing system, and one developer workflow.
BeatAPI is a unified API for AI agents: text, image and video models plus social data and web search behind one API key, with pay-as-you-go pricing.
TypeScript SDK for BlockRun - every frontier model behind one OpenAI-compatible client. Pay with a BlockRun API key or per-call in USDC. Smart routing picks the cheapest capable model. No rate limits.
DSH transport plugin: serves providers whose wire protocols only oh-my-pi's own transport code can speak (Cursor and Devin), by registering its own routes on the dsh-llm seam.
Expose WorkBuddy desktop-app models (CN) through a loopback OpenAI-compatible endpoint, for use as a model provider in ZCode or any OpenAI-compatible client.
Routing infrastructure for AI. One OpenAI-compatible endpoint across 46 providers, with cost caps, 3-layer resilience, caching and guardrails.
Zero-dependency in-process AI Gateway & Control Plane with Cognitive Brain, Multi-Agent Swarm, A2A Handoff Protocol, Universal MCP Ingestion, OpenTelemetry Tracing, Autonomous Agent Loop, Circuit Breaker, Rate Limiting, PII/Security Guardrails, and Unifie
An exact, compiled replacement for the openai SDK's request transform.
Unified AI API for coding tools with clear usage and cost tracking. Single API for Claude, GPT and more; coding-tool setup guides; usage and cost tracking; usage-based pricing.
OpenAI-compatible gateway for Claude, GPT, DeepSeek, Qwen, Kimi & GLM. Flat plans from $19/mo, no per-token bills, 50 free trial requests, no card needed.
One key and one prepaid wallet for agent tools and models — search a tool, check price, run it; pay only for successful calls.
KeepRouter is a managed multi-model API gateway. One account and API key reach the live catalog via OpenAI-compatible /v1/chat/completions, Anthropic-compatible /v1/messages, and route-specific endpoints for embeddings, images, audio, and video. Routing is by explicit public model ID. Billing is prepaid USD pay-as-you-go with no recurring subscription; new accounts can call the published free model before topping up.
Pay-per-token LLM API gateway and OpenRouter alternative: one key for Claude, GPT, Kimi and Grok over OpenAI- and Anthropic-compatible endpoints, priced below vendor list. Each response carries an AWS Nitro attestation you can verify offline; it shows the response bytes came from the upstream provider's TLS endpoint (it does not prove which model ran).
Open-source LLM gateway with one OpenAI-compatible API key for models from OpenAI, Anthropic, Google, DeepSeek and Qwen; agents and apps switch models by changing a string, with automatic failover between providers.
MonkeyAI provides access to multiple AI models through an API relay platform. Developers can use the public English and Chinese documentation to choose the appropriate protocol and configure applications, coding assistants and agent workflows. Available models, supported capabilities and pricing depend on the selected account group. Model pricing, API keys and usage are managed in the account console. API usage is paid; consult the current account pricing before making requests.
01Router provides one API key for leading AI video, image, audio, and chat models. It offers a web studio plus video and image APIs, with model and provider choices to match budget and quality needs.
TheRouter is an OpenAI-compatible AI gateway. Point an OpenAI SDK client, Cursor or Claude Code at one endpoint to reach models from providers such as OpenAI, Anthropic, DeepSeek and Qwen. It adds health-aware routing, automatic failover, real-time token and cost tracking per API key, spending limits and budget alerts, and bring-your-own provider keys. Pricing is pay-as-you-go or BYOK: https://therouter.ai/pricing/ . Quickstart: https://therouter.ai/docs/quickstart/
OmniaKey is a unified AI API gateway for developers, agents and teams. One prepaid API key provides OpenAI-compatible, Anthropic, and Gemini endpoints for text, image, video and audio models, including GPT, Claude, Gemini, DeepSeek, gpt-image, Nano Banana, Seedance and Wan. Pricing is pay-as-you-go with prepaid credits and no subscription. It works with Claude Code, Cursor, Cline and any OpenAI SDK.