Compare AI API aggregators and relays

A Chinese AI API relay based on New API, with multi-model access, a billing console, and a companion creative site.

Website was reachable when last checked

A pay-as-you-go enterprise AI resource platform that aggregates many model APIs and online apps.

Website was reachable when last checked

A publicly listed multi-model API relay whose homepage is a New API console.

Website was reachable when last checked

One API for many models, including GPT, Claude, Gemini, and some Chinese-language models.

Website was reachable when last checked

One API for 1,000+ chat, image, video, audio, and embedding models, aimed at agents and app integrations.

Website was reachable when last checked

An early OpenAI-compatible API service that also offers some Claude endpoints and a reseller mode.

Website was reachable when last checked

A Chinese multi-model API relay.

Website was reachable when last checked

A platform for inference, training, and execution, with hosted APIs and deployment tools.

Website was reachable when last checked

A high-speed inference platform built on wafer-scale chips, with model APIs.

Website was reachable when last checked

A Chinese multi-model API relay often used by individual developers for GPT, DeepSeek, Claude, and similar models.

Website was reachable when last checked

A publicly listed multi-model API relay whose homepage is a New API console.

Website was reachable when last checked

A serverless AI compute platform for deploying and calling models.

Website was reachable when last checked

An enterprise-oriented overseas model API relay/proxy, with multi-protocol access and invoices.

Website was reachable when last checked

Cloudflare’s AI application control plane and gateway for unified calls, caching, rate limits, and observability.

Website was reachable when last checked

A multi-model API relay that emphasizes one key and China-friendly payments.

Website was reachable when last checked

An OpenAI-compatible inference platform for open-source models, billed by usage.

Website was reachable when last checked

A multi-model, multimodal API relay and aggregation service with public docs and pricing.

Website was reachable when last checked

A European AI gateway and LLM router, with one API across about a thousand language models plus OCR, speech, and translation.

Website was reachable when last checked

A generative-media API platform aggregating image and video models, with some language-model APIs as well.

Website was reachable when last checked

Serverless LLM hosting aimed at long-tail open-source models on Hugging Face.

Website was reachable when last checked

A high-performance inference platform for open and custom models, offered as an API.

Website was reachable when last checked

A production inference cloud for AI model-serving APIs.

Website was reachable when last checked

A low-latency inference platform known for its LPU hardware, with model APIs on GroqCloud.

Website was reachable when last checked

Hugging Face Inference Providers: one entry point for models hosted by multiple inference providers.

Website was reachable when last checked

An AI compute and inference platform offering GPU cloud and model APIs.

Website was reachable when last checked

A unified multi-model AI API gateway for GPT, Claude, Gemini, and image APIs.

Website was reachable when last checked

An open-source AI gateway and LLM proxy that calls many providers in one format, with a hosted option.

Website was reachable when last checked

An open-source and hosted LLM API gateway that reaches 40+ model providers through one API.

Website was reachable when last checked

An LLM API relay that emphasizes unified access to multiple models.

Website was reachable when last checked

A multi-model front door for many AI models, with both a web app and an API.

Website was reachable when last checked

An AI cloud and inference platform offering GPU infrastructure and model APIs.

Website was reachable when last checked

A developer platform for 200+ model APIs, an agent sandbox, and GPU cloud.

Website was reachable when last checked

An AI model aggregation platform with one interface to many domestic and international models.

Website was reachable when last checked

A gateway/relay that offers one API for GPT, Claude, Gemini, and 100+ LLMs.

Website was reachable when last checked

A global model-API aggregator used for GPT, Claude, and similar models, including users who need China-friendly payment options.

Website was reachable when last checked

A relay/distribution site for GPT, Claude, Gemini, and image-model APIs.

Website was reachable when last checked

A relay that offers enterprise-oriented ChatGPT and related API distribution.

Website was reachable when last checked

A multi-model API gateway that lets you call models from many labs and inference hosts with one key and one balance, including routing and failover.

Website was reachable when last checked

A global AI API aggregation platform for multi-model access.

Website was reachable when last checked

An inference cloud for AI-native teams, with model-inference APIs.

Website was reachable when last checked

Quora’s multi-model chat and bot platform, with a developer API that spends Poe points.

Website was reachable when last checked

A relay API that describes itself as a direct official-channel reseller, with multi-model access.

Website was reachable when last checked

An AI gateway and control plane for production, with routing, observability, caching, and guardrails.

Website was reachable when last checked

A China-based independent agentic-cloud and MaaS provider for model inference and compute.

Website was reachable when last checked

Run official and community models via API, or deploy your own with Cog; covers image, video, and some language models.

Website was reachable when last checked

An AI gateway and LLM router that advertises access to 600+ models.

Website was reachable when last checked

A high-speed AI inference platform offering hosted APIs for open models.

Website was reachable when last checked

An AI cloud and open-weight model inference platform for developers, with a unified API and compute.

Website was reachable when last checked

An enterprise AI model aggregation gateway and API platform.

Website was reachable when last checked

An enterprise-oriented relay for OpenAI, Claude, Gemini, and similar models.

Website was reachable when last checked

A unified AI gateway in the Vercel ecosystem for routing calls across model providers.

Website was reachable when last checked

YUNWU API, a relay that describes itself as a fast, stable AI API gateway.

Website was reachable when last checked

A unified API for 100+ models including Claude, GPT, and Gemini.

Website was reachable when last checked

Qiniu Cloud’s documentation and access for large-model inference APIs.

Website was reachable when last checked

A one-stop multi-model API relay with usage-based billing and a public pricing page.

Website was reachable when last checked

Infini-AI’s agentic-infra and model platform, offering model APIs and compute.

Website was reachable when last checked

ByteDance Volcengine’s model platform, aggregating Doubao and some third-party model APIs.

Website was reachable when last checked

Baidu AI Cloud’s Qianfan platform for multi-model APIs and applications.

Website was reachable when last checked

A China-based MaaS and inference platform for domestic and open-source models, with a unified API for chat, embeddings, reranking, and images.

Website was reachable when last checked

Alibaba Cloud Model Studio (Bailian), a multi-model application and API platform.

Website was reachable when last checked