Google has announced significant changes to its Gemini AI platform, limiting free users to basic features and prompting a shift towards paid subscriptions for advanced capabilities, including reasoning enhancements and media tools.
Google is tightening access to Gemini for free personal accounts from 9 October 2026, limiting unpaid users to Flash-Lite while reserving Flash and Pro options for subscribers, according to the updated support material. The change marks a further narrowing of the free tier as Google continues to separate basic access from the more capable models and tools bundled with its paid AI plans.
The revised limits are not just about which model appears in the app. Google’s support page says usage is now measured by compute, taking into account prompt complexity, active tools and chat length. Quotas refresh every five hours until the weekly allocation is used up, which means heavier tasks such as longer conversations, tool-heavy prompts and media generation can exhaust allowances more quickly than simple queries. Google’s own documentation also says upgrading unlocks broader model access and additional features inside Gemini Apps.
In Malaysia, Google’s current AI pricing ranges from RM23.99 a month for AI Plus to RM97.99 for AI Pro and RM429.99 for AI Ultra. The price of the top-tier plan was revised earlier this year, when local reports said Google cut the cost of AI Ultra and paired it with 20TB of storage. Under the new structure, AI Plus users keep access to Flash-Lite and Flash, while Pro and Ultra subscribers retain the widest model choice. Google also says AI Pro and AI Ultra customers will be able to use Deep Think, a parallel reasoning feature aimed at more demanding prompts.
The broader direction is consistent with Google’s recent push to segment Gemini more aggressively across price tiers. TechRadar reported that Google has also introduced Gemini 3.1 Flash-Lite for high-volume workloads, while Gemini 3.1 Pro has been rolled out as a preview across subscription levels with a strong emphasis on improved reasoning. Taken together, these changes suggest Google is trying to balance rapid model development with the rising cost of running AI systems, while steering regular users towards paid plans if they want sustained access to the most capable features.
Disclaimer: This content is intended for informational purposes only. Readers are advised to exercise their own judgement, conduct due diligence, or consult a qualified expert before acting on any information provided.
