Businesses adopt smaller language models as AI token costs fluctuate
Artificial intelligence (AI) is getting pricier as token costs keep jumping around, so companies are switching to smaller language models (SLMs) for routine tasks.
Since token prices spike with new model launches and drop when rivals release theirs, it's tough for businesses to figure out if they're really getting their money's worth.
As Mahesh Makhija from EY India puts it, "It's difficult to see the real return on investment from AI because of changing token costs."
Firms set AI budgets, run SLMs
To manage these rising expenses, sometimes as high as $75,000 a month, firms are setting strict budgets and tracking how much value they get from what they spend.
More teams are running SLMs on their own servers for basic jobs, which helps them save cash and stay in control.
As Makhija points out, you don't need a massive model for simple stuff; open-source tools make it easier to keep usage (and costs) under control.