A Surprise in AI Token Usage and Costs

Matthew Berman reacts to a rising usage bill and notes why monitoring token consumption matters.

MBMatthew BermanVortrag · Forward Future
Kanal
Matthew Berman
Veröffentlicht
2. März 2026
Video
0:24
Lesezeit
1 Min.
Original
Englisch
Auf YouTube öffnen
Beim Scrollen schrumpft das Video in die rechte untere Ecke und läuft weiter. Untertitel folgen Ihrer Lesesprache.
LesespracheIn 2 Sprachen verfügbar
Diesen Artikel gibt es noch nicht auf Deutsch, daher wird das Original (Englisch) angezeigt.
ZUSAMMENFASSUNG

In 30 Sekunden

Matthew Berman checks a usage display and says the previous day cost about $130 and the current day had reached $32 by 9:30 a.m. He notes 70 million tokens for the earlier period and 25 million so far that day. The clip does not identify the service, model, pricing or full billing period.

  1. 1An unexpected billBerman reacts with surprise to the reported usage cost.
  2. 2Daily costsHe says the previous day cost about $130 and the current day was already at $32.
  3. 3Token usageHe cites 70 million tokens and 25 million so far that morning; the clip does not specify model or rate.

Das Wichtigste

  1. Berman reports roughly $130 in charges for the previous day and $32 so far the next morning.
  2. He notes token totals of 70 million and 25 million for the periods he describes.
  3. Without the model, rates and full context, these figures cannot establish a general cost per task.
VOLLTEXT

Den bearbeiteten Text nach Kapiteln lesen

Nur das Transkript
01

Watching usage and cost

Berman reviews token totals and charges and says he needs to monitor costs. The clip ends before further detail.

I'm surprised by the cost when I look at the usage display. I say I paid about $130 yesterday and have already reached $32 today, so I need to be careful about spending.

I mention 70 million tokens for the earlier period and 25 million so far today, by 9:30 in the morning. The clip ends as I start discussing costs; it does not identify the service, model or pricing, so the figures should not be generalized to every AI workload.

Mehr aus diesem Kanal

5 Folien14:45EN
Matthew Berman· 29. Okt. 2026Matthew Berman Examines Google’s Gemini 4 Carbon AnnouncementMatthew Berman · Forward FutureMatthew Berman reviews Google’s reported benchmark results, staged rollout and pricing for Gemini 4 Carbon, while noting he has not yet used the model.Foundation ModelsBenchmarksAI cost & performance+2
7 Folien21:11EN
Matthew Berman· 28. Okt. 2026Matthew Berman tests DeepSeek V4.1 Flash: coding, vision and local appsMatthew Berman · Forward FutureA hands-on tour of visual coding, image reasoning, local applications and provider economics reveals both impressive results and visible flaws.Coding-AgentenMultimodal AILocal inference+2
12 Folien34:21EN
Matthew Berman· 27. Okt. 2026Matthew Berman Examines Anthropic’s IPO Filing, AI Growth and GovernanceMatthew Berman, Sam Altman · Forward FutureThe presenter reviews the S-1’s financial metrics, compute strategy and voting structure while questioning the tension between rapid AI development and safety.AnthropicAI SafetyEnterprise-KI+2
0:26EN
Matthew Berman· 25. Okt. 2026Sam Altman on Human Values and Problem-Solving in an AI WorldSam Altman · OpenAIIn a short interview excerpt, Altman considers the enjoyment of solving hard problems and the human role in interpreting results.Human-AI CollaborationFuture of WorkAI Strategy
2 Folien0:36EN
Matthew Berman· 25. Okt. 2026Kimi K3.1 and a Rubik’s Cube Solver DemonstrationMatthew Berman · Forward FutureMatthew Berman shows a scrambled virtual cube being solved and gives the Kimi K3.1 test a favorable assessment.EvalsCoding-AgentenOpen-weight models
2 Folien0:21EN
Matthew Berman· 24. Okt. 2026An AI-Generated Kimi Benchmark Chart Used Invented ScoresMatthew Berman · Forward FutureMatthew Berman asks for a graph of Kimi K3.1 Pro benchmarks, then discovers that the displayed figures are illustrative placeholders.BenchmarksEvalsGenerative media