Changes

Starting in September 2025, Model Studio will launch an LLM inference savings plan to offset token usage fees of online inference for text generation models. Billing is based on the actual consumption and the token-type prices for different LLMs.

The plan offers committed spend tiers of $10, $50, $100, $500, $1000, and $5000, with validity options of 3 months or 6 months.

Thank you for your continued support and trust in Model Studio.