Google’s new experimental Gemini 2.5 model rolls out to free users – Ars Technica

🔍 Summary:

Google has recently made its advanced Gemini 2.5 Pro (Experimental) AI model available to free users, a move that comes shortly after its release exclusively to paying subscribers. This latest model, which is part of the Gemini 2.5 series set to replace the earlier 2.0 version, introduces enhanced capabilities including simulated reasoning to minimize common AI errors. Users can now access most of the new features without needing a Gemini Advanced subscription, although some advanced functionalities like file uploads and the full 1 million-token context window remain limited to paid accounts.

The free version of Gemini 2.5 Pro (Experimental) is accessible via a simple selection from a drop-down menu on the web, with a mobile app rollout expected soon. However, the free access comes with usage restrictions and rate limits, which Google has not fully detailed. During testing, the model allowed for summarizing texts and generating basic code projects before hitting the usage cap, which then required a wait of about three hours for a reset.

Despite these limitations, the release of Gemini 2.5 Pro (Experimental) represents a significant step for Google in its competition with other AI developers like OpenAI. This model offers users a chance to experience some of the latest advancements in AI technology, albeit with certain restrictions aimed at encouraging upgrades to paid subscriptions. As Google continues to refine and expand its AI offerings, further adjustments to token and usage limits can be anticipated.

📌 Source: https://arstechnica.com/gadgets/2025/03/googles-new-experimental-gemini-2-5-model-rolls-out-to-free-users/