AI API cost is the sum of workload units multiplied by model rates, plus retries and failed work. Text often bills input and output tokens separately; image, video, and audio use model-specific units such as generations, resolution, seconds, or characters.
The INFRO view: INFRO gives the text, image, video, and audio models your providers serve one usage and cost-estimation layer, so teams can compare unlike units without losing request context.
Text pricing has two sides
Input includes system instructions, conversation history, retrieved context, tool schemas, and user content. Output includes visible text and, depending on the model contract, billed reasoning or tool-call tokens. Generation-heavy workloads are sensitive to output rates; RAG can be dominated by input.
Estimate with observed percentiles rather than one average prompt. Long-tail contexts can dominate the bill.
Media uses different units
Images may vary by model, count, resolution, quality, and editing inputs. Video commonly depends on duration, resolution, frames, or model tier. Audio can depend on characters, tokens, or seconds. Read the unit next to the rate before comparing providers.
Async job retries and abandoned results still matter if generation completed.
Price per unit is not cost per outcome
Add retries, malformed output, human correction, verification, and orchestration steps. For agents, sum the whole trajectory. For media, include the number of variants needed to produce one usable asset.
The winning model is the one with acceptable quality and the lowest cost per successful task, not necessarily the lowest rate card.
Build a living forecast
Track volume, input and output distributions, model mix, cache effects, retry and fallback rate, media settings, and growth. Recalculate when models, prompts, retrieval, or product behavior changes.
Keep assumptions beside the result so finance and engineering can challenge the same model.
Where INFRO fits
The INFRO pricing page and calculators expose model and workload assumptions, while the request log shows usage, the connection that served it, and estimated provider cost after launch. INFRO charges a subscription and takes no share of model spend; current plans should be checked rather than copied from an old article.
Use a real traffic sample through the quickstart and compare cost per successful outcome. INFRO should be chosen on measured workload economics, not a generic savings promise.
Frequently asked questions
What is an LLM token?
A token is a model's unit for pieces of text. Input and output are often billed at different rates, and the same visible text can tokenize differently across model families.
Why can a cheap model cost more in production?
Retries, longer outputs, invalid responses, extra verification, or lower task success can outweigh a lower per-token price.
Where can I see current INFRO prices?
Use the INFRO pricing page, model catalog, and workload calculators; current pages are more reliable than static figures in an article.
