Skip to content

Fathom Layer / Decision instruments

AI calculators.
Run the numbers.

Estimate model memory, cloud versus local costs, prompt spending and vector storage. Change the inputs and see where your decision changes.

Estimates, not quotes. Every input stays editable.

01 / Find your question

Choose the constraint.

Each instrument solves a different part of a technology decision. Start with the uncertainty in front of you.

02 / Workbench

Change an input.
Watch the answer move.

Use a starting scenario or build your own. Results update as you adjust the controls; the underlying assumptions remain visible.

Instrument 03 / 04

What happens at scale?

Modeled estimate
Explore the sensitivity

Keep a reference. Change one input.

Hold the current estimate, then adjust the controls. Compare how the modeled result changes; the reference stays separate for each instrument.

Current / Live124.10USD

Lowest modeled cost

Across recorded rates · not a quality recommendation

61 tokens (~244 chars)

Prompt text stays in this browser; only the numeric count is shared. Characters divided by four is a rough English heuristic.

1,024 tokens

Cost Matrix (Sorted)

GPT-5.6 Luna$124.10
$0.00124 / call$1241 / 1M
Rate checked 2026-09-16 · official source
DeepSeek V4.1 Flash$124.71
$0.00125 / call$1247 / 1M
Rate checked 2026-09-16 · official source
Gemini 3.5 Flash-Lite$257.83
$0.00258 / call$2578 / 1M
Rate checked 2026-09-16 · official source
Gemini 3.8 Flash$388.57
$0.00389 / call$3886 / 1M
Rate checked 2026-09-16 · official source
DeepSeek V4 Pro$413.56
$0.00414 / call$4136 / 1M
Rate checked 2026-09-16 · official source
Claude Haiku 4.5$518.10
$0.00518 / call$5181 / 1M
Rate checked 2026-09-16 · official source
Gemini 2.5 Pro$1031.63
$0.01032 / call$10316 / 1M
Rate checked 2026-09-16 · official source
Claude Sonnet 5$1036.20
$0.01036 / call$10362 / 1M
Rate checked 2026-09-16 · official source
GPT-5.6 Terra$1241.00
$0.01241 / call$12410 / 1M
Rate checked 2026-09-16 · official source
Gemini 3.1 Pro Preview$1241.00
$0.01241 / call$12410 / 1M
Rate checked 2026-09-16 · official source
GPT-5.6 Sol$2072.40
$0.02072 / call$20724 / 1M
Rate checked 2026-09-16 · official source
Claude Opus 5$2590.50
$0.02590 / call$25905 / 1M
Rate checked 2026-09-16 · official source
GPT-6 Astra$5181.00
$0.05181 / call$51810 / 1M
Rate checked 2026-09-16 · official source
Claude Fable 5.1$5181.00
$0.05181 / call$51810 / 1M
Rate checked 2026-09-16 · official source

Results depend on the entered assumptions. Recorded prices and modeled memory are not live quotes or measured deployment results.

Continue the decision

Keep your inputs in view as you inspect the next part of the decision.

Numeric token counts and request volume carry over. Prompt text stays in this browser.

03 / Starting points

Begin with a real question.

These are editable example inputs. They are not measurements of a deployed system or current vendor quotes.

04 / Read the result

A number is only as useful
as its assumptions.

These calculators model a scenario. Memory estimates include model weights, context cache and overhead; cost estimates depend on the workload and rate values shown in the tool. Actual hardware behavior, provider terms and deployment costs may differ.