Cloudflare AI Gateway vs Not Diamond
Both are llm gateways tools. Here is how they actually differ on price, billing model and deployment.
Cloudflare AI Gateway
Core gateway features free on every Cloudflare plan, running at the edge rather than in one region. Its 2026 unified billing lets you pay OpenAI, Anthropic and Google through a single Cloudflare invoice.
Not Diamond
A model router that recommends rather than proxies - it tells your gateway which model to call and gets out of the way. That architecture means no request-path dependency, which nothing else in this category offers.
| Cloudflare AI Gateway | Not Diamond | |
|---|---|---|
| Category | LLM Gateways | LLM Gateways |
| Our rating | 4/5 | 4/5 |
| Starting price | $0 | Conflicting reports |
| Billing meter | record | No usage metering |
| Free plan | Yes | Yes |
| Free self-hosting | No or paid tier only | No or paid tier only |
| Best for | Teams already on Cloudflare who want caching, rate limiting and cost visibility at the edge with no markup, and who have separate tooling for evaluation. | Teams already running a gateway who want smarter per-request model selection without putting another vendor in the request path, particularly for coding agents where model spend is high and uncontrolled. |
Our verdict on Cloudflare AI Gateway
Cloudflare AI Gateway is the best free managed gateway available, and the reason is that Cloudflare is not trying to make money from it directly. Core features - dashboard analytics, caching, rate limiting - are free on every plan, there is no percentage markup on your tokens, and the free tier includes 100,000 logs a month with the $5 Workers Paid plan lifting that to a million. Compare OpenRouter charging 5.5% of everything you spend forever, and the difference at volume is substantial. Running at the edge is a genuine architectural advantage too, putting the control plane near your users rather than in one region. The 2026 addition of unified billing, letting you pay for OpenAI, Anthropic and Google AI Studio usage through a single Cloudflare invoice, removes procurement friction that is more annoying in practice than it sounds. Two honest limits. The metering is logs, not requests, so that is the number to model. And observability stops at request metrics and cost - there is no structured evaluation, so this is one component of a stack rather than the stack.
Full Cloudflare AI Gateway review →Our verdict on Not Diamond
Not Diamond is architecturally the most interesting thing in this category because it declines to be a gateway. Integrations are stack-agnostic through its API, and recommendations are executed in your own model gateway and harness - so it tells you which model to call and gets out of the way. Every other router in this category is a proxy, which means adopting it creates a request-path dependency and a new availability risk. Not Diamond has neither. It composes with LiteLLM, Vercel, Cloudflare or your own proxy rather than replacing any of them, which makes it unusually low-commitment to try. The product is a pre-trained Chat router, a Code router aimed at coding-agent cost currently in early access, and custom routers trained on your own data. That last one matters most, because the weakness of generic routing is that your workload is not the benchmark it was tuned on. The unsatisfying part is pricing, where sources directly contradict each other, and the fact that the most differentiated capability is still early access.
Full Not Diamond review →Frequently Asked Questions
What is the main difference between Cloudflare AI Gateway and Not Diamond?
Cloudflare AI Gateway: Teams already on Cloudflare who want caching, rate limiting and cost visibility at the edge with no markup, and who have separate tooling for evaluation. Not Diamond: Teams already running a gateway who want smarter per-request model selection without putting another vendor in the request path, particularly for coding agents where model spend is high and uncontrolled. Both sit in LLM Gateways, so the decision usually comes down to billing model and deployment rather than raw capability.
Which is cheaper, Cloudflare AI Gateway or Not Diamond?
It depends entirely on your workload shape, because they meter differently - Cloudflare AI Gateway bills on record and Not Diamond bills on no usage metering. Published starting prices are $0 and Conflicting reports respectively, but those numbers are not comparable until you apply them to the same traffic. Use our cost calculator to model both against your own request volume and span count.
Can I self-host Cloudflare AI Gateway or Not Diamond?
Cloudflare AI Gateway: No or paid tier only. Not Diamond: No or paid tier only. Free self-hosting means no licence fee, not no cost - you still own the infrastructure, upgrades and on-call.