Task: screen one wallet address against the OFAC SDN list and the UK, EU, Canadian and US Consolidated Screening lists. Input: one address that is on the OFAC SDN list. Measured 2026-10-28.
| Requests | Size | Time | Tokens if read into a model | |
|---|---|---|---|---|
| By hand: download the five official list files and parse them | 5 requests | 97.0 MB | 36 s | about 32 million tokens |
One call: one call to a prepared index of the same five lists (/sanctions/check) | 1 call | 4.1 KB | 1.4 s | about 1,300 tokens |
| Tokens | Cost | |
|---|---|---|
| By hand: the raw content read into a model | about 32 million tokens | about $63.94 in model input tokens |
| One call: the price of the call | $0.005 | |
| One call: the answer's own tokens read into a model | about 1,300 tokens | about $0.0026 |
| One call, together | about $0.0076 |
Tokens are priced at $2.00 per 1 million input tokens (OpenAI gpt-4.1, standard tier, text input tokens). Source: https://developers.openai.com/api/docs/pricing, read 2026-10-28. The price of the call is the amount in the route's own payment offer when it was measured (2026-10-28). Only input tokens are counted, not the model's own output. Prices change; these figures hold for those dates, and another model costs more or less. A local parser spends no tokens but needs code and upkeep, and some by-hand sizes are larger than a model's context window, so the by-hand figure is the price of the tokens, not a workable procedure.
| Source | Address | HTTP | Size | Time | Tokens |
|---|---|---|---|---|---|
| OFAC SDN | www.treasury.gov/ofac/downloads/sdn.xml | 200 | 29.3 MB | 16.6 s | 10,291,351 |
| UK Sanctions List | sanctionslist.fcdo.gov.uk/docs/UK-Sanctions-List.xml | 200 | 22.1 MB | 1.3 s | 6,496,084 |
| EU Financial Sanctions | webgate.ec.europa.eu/fsd/fsf/public/files/xmlFullSanctionsList_1_1/content | 200 | 25.9 MB | 7.0 s | 8,615,934 |
| Canada SEMA/JVCFOA | www.international.gc.ca/world-monde/assets/office_docs/international_relations-rela | 200 | 2.9 MB | 7.0 s | 1,043,693 |
| US Consolidated Screening List | data.trade.gov/downloadable_consolidated_screening_list/v1/consolidated.cs | 200 | 16.8 MB | 4.3 s | 5,524,430 |
A local parser reads files without spending tokens but needs code and upkeep; a cached copy avoids the download but goes stale; accuracy, completeness and price of any answer are separate questions.
Repeat: Re-measured weekly by agents/savings_facts.py; the harness is ops/savings_measure.mjs. Raw numbers as JSON: ?format=json.
Published by P0, run by dev Mike. An open index of the machine-to-machine (agent) economy. Terms · Who runs P0 · All measurements