Google restricts advanced Gemini models to paid subscribers

Starting October 9, free users of Google's Gemini will only have access to the Flash-Lite model, while AI Plus subscribers will lose access to the Pro model. This change marks a shift similar to OpenAI's recent restrictions, as the high cost of running advanced AI models forces companies to limit access. According to Google's updated support page, free users and AI Plus subscribers will no longer be able to use the Flash or Pro models unless they upgrade to a paid plan.

Google's changes also introduce effort settings, Low, Medium, and High, which allow users to balance response quality against their usage allowance. Higher effort levels may produce better results but consume more tokens, adding complexity to the user experience. This approach mirrors ChatGPT's model picker, where users can adjust effort levels, though some find the labels abstract and confusing.

Industry-wide, AI companies are struggling with the financial burden of running powerful models. OpenAI recently reduced advanced-model allowances on some paid plans, leading to user dissatisfaction. Meanwhile, Anthropic took a more generous stance by increasing usage limits for its Claude Opus 5.5 users, though complaints about reaching limits persist. The trend reflects a broader strategy in Silicon Valley to offer services cheaply initially and then raise prices as demand grows.

For Gemini users, the upcoming changes mean that accessing the most advanced models will now depend on subscription tiers. Flash-Lite remains available to all, Flash is restricted to paid tiers, and Pro is exclusively for AI Pro and AI Ultra subscribers. As AI models grow smarter, their accessibility will increasingly hinge on how much users are willing to pay.