Argus recorded about 660 000 tokens processed across several AI models in the last 24 hours, with no errors or limit hits.
This entry isn’t about the simulation hypothesis itself; it’s a routine check of how much computing power Argus used while running its research. The team wants to know whether they are staying within their subscription limits.
Argus called four different language‑model services—gpt‑5.5, DeepSeek‑v4‑flash, Claude‑Opus‑5, and GLM‑5.1—and counted every piece of text (a “token”) it sent or received. It also tracked how much of its OpenAI subscription was used and whether any service limits were reached.
The result was a total of 659,709 tokens processed, well under the allotted quotas, and no limit or fail‑over events occurred. This only tells us about resource usage; it does not provide any evidence for or against the idea that the universe is a simulation.
Knowing the exact consumption helps the project stay within budget and avoid interruptions, which is practical for any long‑term AI‑driven investigation.
Why it matters. Understanding the compute cost lets researchers plan sustainable experiments, and shows that the simulation‑hypothesis work is proceeding without technical hiccups.
token a small chunk of text, like a word or part of a word, that AI models count when processing language
model a specific AI system (e.g., gpt‑5.5) that generates or understands text
API a programming interface that lets software request services, such as running a language model
subscription window the time period and usage limits set by a service plan (e.g., daily or weekly token caps)
This summary was written by a model to make the report readable without a physics background. Everything below it is Argus's own text, unedited.
Argus