Cloudflare AI Gateway
A proxy layer that routes, caches, rate limits and logs requests to AI model providers from one endpoint.
How the 7.0 was reached
- Quality of output6/10
Independent ComputeSDK benchmarks show warm first-token latency around 455-478ms through the gateway; no broader quality evidence. computesdk.com
- Ease of use8/10
Free start with no credit card and a docs site at developers.cloudflare.com. cloudflare.com
- Pricing value7/10
Free plan, Pro from $20/month and Business from $200/month; the free plan has hard limits and users report unexpected bills. cloudflare.com
- API and integration quality9/10
REST API and Workers Bindings, plus many providers including OpenAI, Anthropic, Bedrock and Azure. developers.cloudflare.com
- Solves the problem it claims to solve5/10
Routing and logging are documented, but the only fit evidence is a vendor page mentioning surprise bills, free-plan throttling and unclear data leakage. blog.cloudflare.com
7.0/10 is the mean of the 5 criteria that apply. Scored on the five-criteria rubric v1, 29 September 2026. How scoring works
Pricing
- For personal or hobby projects that aren't business-critical
- Professional sites that aren't business-critical
- For small businesses operating online
- Billed annually
- Mission-critical applications
Key Features
- Dynamic routing across model providers
- Caching, spend limits and rate limiting
- REST API and Workers Bindings
- Integrations with OpenAI, Anthropic, Google, Azure OpenAI, Bedrock, Groq and DeepSeek
- Usage, billing and log tracking
Pros & Cons
Pros
- Broad provider coverage and both REST and Workers integration
- Free start with no credit card
- Runs on Cloudflare's global network
- Spend limits and rate limits help control cost
Cons
- The plan tiers on the pricing page are Cloudflare-wide, so AI Gateway-specific limits are unclear
- The free plan has hard limits, and users report unexpected bills
- Only a vendor-side page backs the fit claim, mentioning throttling and unclear data leakage
- Independent latency figures of about 455-478ms warm first-token are not fast
It suits teams already on Cloudflare that want one place to route and log calls to several model providers. Anyone who needs firm cost predictability or clear data handling terms should read the limits closely first. The 7 rating reflects good integration but thin evidence on fit.
Try Cloudflare AI Gateway →Added to scored.tools on
Competitors to Cloudflare AI Gateway
Other tools in the automation category worth comparing.
Ollama
8.6/10Run and deploy large language models locally with enterprise-grade cloud scaling options.
n8n
8.4/10Open-source workflow automation with a visual builder and self-hosting option.
Dify
7.6/10Build production-ready agentic workflows, RAG pipelines, and autonomous agents at scale.
Zapier
7.2/10No-code automation platform connecting 9,000+ apps with AI-powered workflows and agents.