DeepSeek API gateways

Flow7's prepaid multi-model API gateway provides Responses, Chat Completions, and Anthropic Messages compatibility, request spend limits, locked price versions, and completed-response charge receipts. Public model-family labels do not establish verified upstream supplier or model-weight provenance.

A multi-model routing platform for developers; switch models by changing the base_url with the OpenAI SDK, ideal for teams comparing the quality and price of many models at once.

Vercel's official AI Gateway, emphasizing unified model calls, failover, caching, and observability; ideal for apps already deployed on Vercel.

A tool and API aggregation platform covering text, image, video, and audio AI capabilities; ideal for individuals and small teams wanting one-stop access to many models.

A common API relay service among Chinese users; the site highlights aggregation of OpenAI, Claude, and Gemini plus multiple top-up methods.

A China-based model API platform offering DeepSeek, Qwen, image, speech, and vector capabilities; ideal for developers who prioritize domestic models and low-latency nodes.

A multi-model API aggregation platform; the docs list an OpenAI-compatible interface, model list, and image/audio/video APIs, ideal for overseas developers wanting unified access.

An enterprise-oriented AI Gateway focused on a unified interface, routing, fallback, observability, and logging; ideal for teams needing to govern multi-model calls.

An open-source LLM gateway/proxy layer that unifies many providers into an OpenAI-style interface; ideal for teams that want to self-host a relay and control their own keys and logs.

An open-source project for managing and distributing OpenAI-style interfaces; commonly used to self-host an API relay with channel management, quotas, and token grouping.

An open-source multi-channel gateway evolved from the One API ecosystem, with more model channels, billing features, and operational config; ideal for webmasters or teams self-hosting a unified API entry point. The repository has moved from Calcium-Ion to the QuantumNous organization, with the old address redirecting.

An overseas open-model inference platform covering Llama, Qwen, DeepSeek, and more, also offering fine-tuning and serverless inference.

An overseas enterprise-grade open-model inference platform covering Llama, Qwen, DeepSeek, and more, with function calling, JSON mode, and fine-tuning.

A multi-model API aggregation service compiled from public info; it shows many integrated models and support for WeChat, Alipay, and crypto top-up.

A multi-model aggregation API service compiled from public info; it shows moderate pricing, acceptable stability, and WeChat/Alipay top-up.

A unified LLM API gateway for China-based developers; public info shows support for Claude, GPT, Gemini, DeepSeek, and adaptation for tools like Cursor, Claude Code, and Cline.

A China-based multi-model relay; public info shows support for ChatGPT, Claude, Gemini, DeepSeek, and emphasizes China direct connection.

DeepSeek's official API platform, focused on cost-effective Chinese, code, and reasoning models; ideal for apps prioritizing domestic models.

Alibaba Cloud's LLM service platform, offering the Tongyi Qwen series and various third-party models; ideal for teams needing a cloud account, invoices, and enterprise-grade service.

Baidu AI Cloud's Qianfan LLM platform, covering ERNIE, DeepSeek, and some open models; ideal for the Baidu Cloud ecosystem and enterprise-service scenarios.