In 30 seconds
Theo recounts testing Fable and switching to usage-based billing after reaching a subscription limit. He says the run spent $100 in roughly eight minutes, and a second subscription reached its five-hour session limit two hours later; these are his experience at the time, not a general price comparison.
- 1A costly switchTheo says usage-based billing consumed $100 in about eight minutes.
- 2A second session limitHe got another subscription and says its five-hour session limit was exhausted in two hours.
- 3Work left unfinishedOne workflow hit both limits, and Theo stopped rather than incur additional costs.
Takeaways
- Theo says his usage-based run spent $100 in about eight minutes.
- He then used another subscription and reached its five-hour session limit in roughly two hours.
- The workflow remained unfinished, and he chose not to spend more to continue it.
- Subscription limits and usage-based charges can vary by plan and workload.
Read the edited text by chapter
Usage-based billing and limits
Theo describes switching billing methods and quickly spending $100. He then reaches the session limit on a second subscription.
I switched to usage-based billing after hitting my subscription limit, and it turned out to be a mistake for this run: it spent $100 in about eight minutes. My subscription usage limit seemed like a better deal for my workload, so I got another subscription and hit its five-hour session limit in another two hours.
An unfinished workflow
The task remained incomplete after the limits were reached. Theo decided not to continue spending on it.
One run, one workflow and one prompt were enough to exhaust the session limits on two subscriptions. The task still wasn't finished when I went to bed, and I didn't want to spend potentially another thousand dollars on it. It was a rough experience.
More from this channel
38:47EN
Theo - t3․gg· Oct 28, 2026Theo’s Long-Running Agent Experiments: Verification, Compute and CostTheo Browne · Ping LabsA skateboard game and two compiler projects show how repeated checks and flexible compute can sustain agent work—and how expensive those loops can become.
36:07ENTheo - t3․gg· Oct 27, 2026Theo on Token Costs, Coding Agents, and the Price of AI CapabilityTheo Browne · Ping LabsA viral estimate of Lauren's AI spending prompts a closer look at cache pricing, benchmarks, and the changing work of coding agents.
30:37ENTheo - t3․gg· Oct 26, 2026Theo’s Three Skills for Reviewing and Shipping with Coding AgentsTheo Browne · Ping LabsHow sol-loop, full-send and full-stop encode review, verification and different levels of merge authority.
0:43ENTheo - t3․gg· Oct 23, 2026Theo Distinguishes Security Vulnerabilities from Active AttacksTheo Browne · Ping LabsA short explanation of why disclosed software flaws and reported attacks are different kinds of security news.
29:22ENTheo - t3․gg· Oct 22, 2026Theo on Security Vulnerabilities, Stolen API Keys and AITheo Browne · Ping LabsA look at the gap between vulnerability disclosure and patching, and a speculative account of why stolen inference keys may be valuable.
31:13ENTheo - t3․gg· Oct 20, 2026Theo on Why Smarter Models Help More Than DevelopersTheo Browne · Ping LabsFrom game modifications to compiler experiments, Theo revisits his assumptions about who benefits from more capable AI.