Global – October 24, 2026 -- Cloudflare is pitching Workers AI as a serverless inference platform that bills only for usage, at $0.011 per thousand neurons, and runs 50+ models across 200+ cities.
Cloudflare targets idle GPU spend with usage-based billing
Cloudflare argues that inference demand is spiky, unlike training. It cites average GPU utilization of only 20-40%, with one-third of organizations below 15%. Workers AI removes hardware commitments, GPU management and capacity planning, so customers pay only for inference they run.
The pricing page lists 10,000 neurons per day alongside the $0.011 per thousand neurons rate.
Platform exposes Kimi K2.6, GLM 4.7 Flash, GPT-OSS-120B and Llama 4 Scout through one API
The catalog includes:
- Kimi K2.6, a vision and agentic tool-calling model
- GLM 4.7 Flash, a multilingual agent model
- GPT-OSS-120B, aimed at coding and debugging
- Llama 4 Scout, a generalist model
Developers call models through a single endpoint, via Workers code, REST or any OpenAI SDK. Cloudflare says the platform handles provisioning, scaling and latency optimization automatically.
Supported workloads include LLMs, image generation, real-time speech-to-text and embeddings. The service runs on infrastructure that Cloudflare says powers 20% of the Internet.
Shopify cites simplicity at scale
Shopify says Cloudflare helps it achieve complex outcomes in a simple way that it can scale and maintain.