BusinessThe Decoder· Jun 30, 2026
OpenAI Reportedly Halves ChatGPT Inference Costs
OpenAI has reduced inference costs for its AI models by more than half, according to a report from The Information. The optimizations were implemented in ChatGPT and lowered the number of Nvidia GPUs required to just a few hundred during operation. These changes focus on improving efficiency for guest users accessing the service.
Key points
- →Inference costs for OpenAI models cut by over 50 percent
- →Optimizations applied directly to ChatGPT operations
- →Nvidia GPU usage reduced significantly at times
- →Details reported by The Information
Mentioned
OpenAIChatGPTNvidiaThe Information
Related stories
Etched Reaches $5 Billion Valuation with $1 Billion in AI Chip ContractsTechCrunch · Hardware→OverDrive CEO Sees AI as Key Focus for Ebook AppThe Verge · Business→Meituan Trains 1.6 Trillion Parameter Model on Chinese ChipsThe Decoder · Models→San Francisco AI Boom Increases Housing Costs for High-Earning Tech WorkersThe Decoder · Business→Amazon Forms $1 Billion Organization for AI Agent DeploymentsTechCrunch · Business→Lawyer Bill Savitt Defeats Elon Musk in Court TwiceThe Verge · Business→
This is an original summary by Dhanasvi's agents based on The Decoder's public feed. For the complete article, visit the original source. Trademarks and article copyright belong to their owners.