Workers AI
Serverless GPU inference on Cloudflare’s global network.
Run open models — LLMs, embeddings, speech, image and Cloudflare’s own Clef decision models — on serverless GPUs across Cloudflare’s network, called from Workers or a REST API and billed per use.
- Vendor
- Cloudflare
- Category
- Inference & model hosting
- Pricing
- Freemium
- Open source
- No
- Platforms
- API, Workers
- Launched
- 2023
- Website
- developers.cloudflare.com/workers-ai
Features
- Serverless GPUs at the edge
- Open-model catalog
- Clef decision models
Best for
- High volume / low cost
- Real-time / low latency
Works with
- Granite 4.0 H Micro — IBM. Tiny hybrid Mamba-2/Transformer model for long-input enterprise tasks.
- Clef — Cloudflare. Cloudflare’s open-source 27B multimodal decision model for high-precision decisions.
- Clef-flash — Cloudflare. The fast 9B Clef for latency-critical, hot-path decisions.
- Clef Omni — Cloudflare. Clef’s mixture-of-experts member, deciding over text, images, audio and video.
More inference & model hosting
- Ollama — Ollama. Run open models locally with one command.
- vLLM — vLLM project. High-throughput LLM serving engine.
- llama.cpp — ggml. LLM inference in C/C++ on any hardware.
- Amazon Bedrock — AWS. Foundation models on AWS.
- Vertex AI — Google Cloud. Google Cloud’s AI platform.
- LM Studio — Element Labs. Desktop app to discover and run local LLMs.
- Azure AI Foundry — Microsoft. Build and run AI apps on Azure.
- Groq — Groq. Ultra-fast inference on custom LPUs.