AI Costs Fall as Data Center Backlash Looms
AI inference costs have fallen so far this year despite rising server-rental prices, with Artificial Analysis finding that GPT-6.1 Sol costs about one-fifth as much per task as GPT-6 Astra at a high reasoning setting. Anthropic has promoted similar efficiency gains in its latest models, though that claim comes from the vendor, while buying or renting Nvidia hardware has become substantially more expensive. The Information reports that post-election restrictions on data-center expansion could raise operating costs, but the scale and timing of any effect remain uncertain.
Continued efficiency gains could sustain lower AI operating costs even as more expensive hardware and data-center delays increase infrastructure costs, a trend that matters as enterprises weigh AI spending.
This source provides an excerpt. Continue reading on the original site.
#OpenAI#GPT-6.1 Sol#GPT-6 Astra#Anthropic#Nvidia#Amazon Web Services#Blackstone#Anthropic IPO