No. 001
Inference Economics
$1 of inference to 1 million people is $1 million.
Why AI companies will eventually need to decide which workloads really require frontier intelligence, and which don't.
Publishing soon
Spend intelligence deliberately.
AI products don’t need the smartest model for every request.
The difficult part isn’t finding a cheaper model.
It’s knowing where you can make that trade without breaking the product.
Inference economics and AI infrastructure
Which parts of your product actually need frontier intelligence?
01 · Economics
$1 of inference
× 1,000,000 users
= $1,000,000
A two-cent model call looks irrelevant.
At scale, request-level decisions become infrastructure decisions.
Companies need to understand:
- 01what each product feature costs to serve
- 02which workloads actually require frontier models
- 03whether repeated context is being cached effectively
- 04whether smaller or open-weight models preserve acceptable quality
- 05when batching changes the economics
- 06when self-hosting beats API pricing at realistic utilization
- 07what latency and throughput look like under real concurrency
02 · Three dimensions
Economics
Where inference spend actually goes and which interventions produce measurable savings.
Performance
How model, runtime, hardware, batching, caching, concurrency, and utilization change the operating point.
Quality
Whether a cheaper execution path remains good enough for the workload it actually performs.
The cheapest model is not necessarily the cheapest system.
03 · Direction
From measurement to execution.
Today, Enough is focused on measuring and understanding inference workloads.
The larger question is:
Given this workload and these constraints, where should it execute?
MeasureEvaluateChooseExecuteObserve
Subject to
- cost
- latency
- quality
- privacy
- hardware availability
- reliability
This is the direction of the work, not a platform that exists today. The measurement comes first.
04 · Writing
Research notes
More experiments and benchmark results coming.
05 · Work with me
Have a meaningful inference bill?
I’m looking for real production workloads to analyze.
If your company is spending meaningfully on OpenAI, Anthropic, Gemini, hosted open-weight models, or self-hosted inference, I’d like to understand where the money is going and which questions are worth investigating.
I’m particularly interested in anonymized usage exports.
No AI transformation workshop. Just the workload, the bill, and the engineering.