Cloudflare Workers AI
An edge AI inference platform provided by Cloudflare that allows developers to run large model inference on Cloudflare's global network. Usage is metered in Neurons, and a daily free allowance is provided.
Core Facts
Cloudflare Workers AI has multiple embedding models built in, with all model inference completed by Cloudflare Workers AI.
Muse Wonder offers three Meta API request formats (Responses, Chat Completions, Messages), Muse Code, and Vercel AI Gateway and Cloudflare Workers AI integrations.
Timeline (last 90 days)
The free Providers that 9router supports connecting to include OpenCode Free (no authentication required), Kiro AI (OAuth, about 50 quota per month, supporting Claude 4.5, GLM-5, MiniMax), Vertex AI (OAuth, $300 credit for new GCP accounts), NVIDIA NIM (API Key), and Cloudflare Workers AI (API Key).
Vendors such as Gemini, Groq, NVIDIA NIM, OpenRouter, Cloudflare Workers AI and Cerebras provide OpenAI-compatible chat format endpoints.
Muse Wonder offers three Meta API request formats (Responses, Chat Completions, Messages), Muse Code, and Vercel AI Gateway and Cloudflare Workers AI integrations.
Vercel AI Gateway and Cloudflare Workers AI offer the meta/muse-wonder-1.0 service, and whether an independent quota is obtained depends on how the gateway is deployed.
MainBrella's dialogue reasoning is handled by Llama 3.3 70B, invoked through Cloudflare Workers AI.
Cloudflare Workers AI is billed by neurons or call volume and offers out-of-the-box open-source models such as Llama 3.3 70B, invoked via API, but model selection, fine-tuning and context window are limited by the platform.
Cloudflare Workers AI is a serverless inference service provided by Cloudflare on its global edge nodes, billed by Neurons; @cf/qwen/qwen2.5-7b-instruct is its hosted instruction-tuned version of Qwen 2.5 7B, from the same source as the weights used locally.
When BillCraft AI's backend cannot access the local Ollama, it switches to Cloudflare Workers AI (@cf/qwen/qwen2.5-7b-instruct) as cloud fallback inference.
BillCraft AI's privacy promise contains tension: the product's core promise is privacy-first, but the demo environment actually uses Cloudflare cloud inference, and the author did not explain in detail the trigger conditions of the switching logic or the user notification mechanism.
Cloudflare Workers AI is billed by neuron, with pricing that is relatively fixed and predictable.
2 more timeline events
All Facts (14)
Cloudflare Workers AI has multiple embedding models built in, with all model inference completed by Cloudflare Workers AI.
65%Partially VerifiedMuse Wonder offers three Meta API request formats (Responses, Chat Completions, Messages), Muse Code, and Vercel AI Gateway and Cloudflare Workers AI integrations.
70%UnverifiedThe free Providers that 9router supports connecting to include OpenCode Free (no authentication required), Kiro AI (OAuth, about 50 quota per month, supporting Claude 4.5, GLM-5, MiniMax), Vertex AI (OAuth, $300 credit for new GCP accounts), NVIDIA NIM (API Key), and Cloudflare Workers AI (API Key).
50%UnverifiedVendors such as Gemini, Groq, NVIDIA NIM, OpenRouter, Cloudflare Workers AI and Cerebras provide OpenAI-compatible chat format endpoints.
50%UnverifiedVercel AI Gateway and Cloudflare Workers AI offer the meta/muse-wonder-1.0 service, and whether an independent quota is obtained depends on how the gateway is deployed.
50%UnverifiedMainBrella's dialogue reasoning is handled by Llama 3.3 70B, invoked through Cloudflare Workers AI.
50%UnverifiedCloudflare Workers AI is billed by neurons or call volume and offers out-of-the-box open-source models such as Llama 3.3 70B, invoked via API, but model selection, fine-tuning and context window are limited by the platform.
50%UnverifiedWhen BillCraft AI's backend cannot access the local Ollama, it switches to Cloudflare Workers AI (@cf/qwen/qwen2.5-7b-instruct) as cloud fallback inference.
50%UnverifiedCloudflare Workers AI is a serverless inference service provided by Cloudflare on its global edge nodes, billed by Neurons; @cf/qwen/qwen2.5-7b-instruct is its hosted instruction-tuned version of Qwen 2.5 7B, from the same source as the weights used locally.
50%UnverifiedBillCraft AI's privacy promise contains tension: the product's core promise is privacy-first, but the demo environment actually uses Cloudflare cloud inference, and the author did not explain in detail the trigger conditions of the switching logic or the user notification mechanism.
50%UnverifiedCloudflare Workers AI is billed by neuron, with pricing that is relatively fixed and predictable.
50%UnverifiedClef is an open-weight model that can be called via Cloudflare Workers AI or run locally with Ollama.
50%UnverifiedIn the JS/TS AI development space, AI SDK faces competition from LangChain.js, Microsoft Semantic Kernel, Cloudflare Workers AI, and others.
50%UnverifiedAI 模型的流式回复通过 Server-Sent Events（SSE）或 WebSocket 实现 token 级推送，OpenAI、Anthropic 等主流 API 均采用这一方式
50%