Google Cloud does not expect enterprises to standardise on Gemini alone, believing the future of enterprise AI will be built on a mix of frontier, open, and third-party models. Richard Seroter, Chief Evangelist at Google Cloud, stated during a media roundtable at Google I/O Connect in Bengaluru that "the world is probably going to be multi-model." With governments influencing access to frontier AI models through export controls and regulations, Seroter noted no company can control decisions made by competitors or governments, referencing Anthropic’s Fable 5 and Mythos 5. Despite competition from lower-cost Chinese AI models like GLM 5.2 and DeepSeek V4, Seroter said Google has no intention of forcing customers to use Gemini, instead offering competing models through Model Garden on Google Cloud infrastructure. Model Garden provides access to hundreds of models, including Google’s Gemma family and models from Anthropic, Meta, Mistral AI, DeepSeek, Kimi, Z.ai, and thousands from Hugging Face. Google is expected to release Gemini 3.5 Pro soon. Seroter also highlighted small language models like Gemma and Gemini Nano for local or on-device use, and acknowledged that token pricing for India has remained relatively stable.