Inference vs training

Training is the one-time event.
Inference is the lifetime opex.

Training a frontier model costs tens of millions of dollars. Serving it to a billion agents, continuously, 24/7, costs a hundred times more — every year, for the life of the model. That's why 80–90% of every AI system's lifetime compute spend is inference.

The lifetime cost of an AI system

Same model, same vendor, real numbers. Training is the spike. Inference is the floor.

Dimension Training Inference
When it runs Once per model version — days or weeks Every request, 24/7, indefinitely
Accounting treatment One-time capital event / R&D Recurring operating expense
Per-run cost $10M – $500M+ for a frontier run ¢ to $ per request, $M to $B per year
Share of lifetime compute 10 – 20% 80 – 90%
Cost trend Climbing at the frontier (bigger models, more tokens) Per-token cost falling 9× to 900× per year
Primary cost driver Model size and dataset scale Token volume × request latency
Workload pattern Bursty, batch, predictable Continuous, bursty, latency-bound, geo-distributed
Geography One cluster, one site, one power deal Many regions, edge, on-prem, near user
Buyer inside the enterprise Head of AI / R&D Head of Engineering, Product, CX, FinOps

Two different products, two different racks

The economics of training and inference are different enough that the rack profile is different too.

🏋

Training rack profile

Maximizes per-GPU FLOPs and interconnect bandwidth for large all-reduce operations. Runs one job, saturates the fabric, then idles. Optimized for batch throughput, not latency. Liquid-cooled, megawatt-class.

Inference rack profile

Maximizes tokens-per-dollar-per-watt at target latency. Runs millions of requests concurrently, dynamically batched. Optimized for KV-cache capacity, prefill/decode disaggregation, and per-request SLA. Often smaller, denser, more numerous.

🔁

The overlap is intentional

Modern AI factories run both on the same rack platform — the NVL72 is explicitly sold as "training and inference in one." The fact that it can do both is a feature, not a product identity. The product identity is "the rack."

📈

Why this matters for the domain

The product identity is the inference rack, because inference is 80–90% of the spend. Even a rack that does both will be marketed, named, and remembered for its inference profile. That's the brand. That's the URL.

How the inference budget compounds

A worked example: an enterprise that trains a $30M frontier model in 2026. How much does it spend serving that model over its lifetime?

$30M

Training event

One frontier training run, 2026. Single line item. Capitalized or expensed depending on accounting.

$5–15M

Year 1 inference

Initial rollout, mostly internal users, copilots, a handful of agent pilots. Typical 2024–2025 baseline.

$50–150M

Year 2 inference

Department-scale agents, customer-facing products, embedded AI. Typical 2026–2027 enterprise projection.

$500M–1.5B

Years 3–4 inference

Enterprise-wide agent fleets, AI as the primary interface. Typical 2028–2029 projection at large enterprises.

10×

Model retraining

The model gets retrained every 6–18 months at the frontier. So training spend reappears — but still as ~10–20% of total.

~100×

Lifetime ratio

Over 4 years, the typical large-enterprise AI system spends ~10× more on inference than training. Some models will hit 100×.

"Inference now consumes the majority of an AI system's lifetime compute, with industry analyses putting it at roughly 80 to 90% of total compute dollars over a model's lifecycle versus 10 to 20% for training."

Telnyx, AI training vs inference: the 2026 economics split

"The AI inference market will grow from $106 billion in 2025 to $255 billion by 2030, with a 19.2% compound annual growth rate. Gartner projects that 55% of AI-optimized IaaS spending will support inference workloads in 2026, reaching over 65% by 2029."

Introl / Gartner, AI Inference vs Training Infrastructure economics

80–90% of every AI dollar is spent on inference. Own the address.

InferenceRacks.com — the literal phrase, the dominant workload, the unit of the market.

Acquire the domain →