Cloudflare Workers AI 提供免费 API — 在边缘运行 LLM 而无需支付 OpenAI

Cloudflare Workers AI offers a free API that allows users to run large language models (LLMs) at the edge without incurring costs from OpenAI. The service supports various models, including Llama 3, Stable Diffusion, and Whisper, and can be accessed via a simple curl command. The free tier includes 10,000 neurons per day, sufficient for approximately 100-500 requests, and supports functionalities such as text generation, image generation, speech-to-text, translation, and embeddings. Cloudflare's global network ensures edge deployment with no cold starts, keeping models always warm and providing production-ready AI inference capabilities.