Home›Global Markets›Trade & Tariffs›Open-weight AI models cut LLM inference prices, easing…

Open-weight AI models cut LLM inference prices, easing industry fears

Inference costs per million tokens fell from above $2 in early June to $1.2 this week, according to Silicon Data’s index.

Reported · Written up by our desk · Trade & Tariffs

Breakthroughs in cheaper Chinese open-weight AI models have rattled some US investors, but analysts argue the price drop will expand demand for AI systems and accelerate adoption across the industry, according to SCMP Economy. SCMP Economy notes that AI companies have reduced pricing in recent weeks, with large-language model inference costs per million tokens falling from above US$2 at the start of June to US$1.2 this week, based on Silicon Data’s LLM Token Expenditure Index. The index tracks both frontier providers and open-weight platforms. Silicon Data said competition is increasing and prices are falling, adding that lower costs should benefit consumer and enterprise users of AI, including agent-style applications. The outlet frames the dynamic as a near-term concern for investors, with a longer-term upside for broader and faster AI adoption.

SCMP Economy attributes the commentary to Silicon Data, which said the shift should promote wider use of AI (agents) and speed up deployment.

More like this

Sources