Urgent.News

What's breaking now, across thousands of outlets.

AI

Free Gemini app users now only have access to ‘Auto’ models

As announced last week , Google is rolling out a big change to the Gemini app experience for users on the free plan. more…

Free Gemini app users now only have access to ‘Auto’ models

Google has announced a major update to the Gemini app experience for free users, limiting their model selection options. Free users without a plan can no longer choose their preferred model; they are now only able to select the "Auto" model with "Gemini 3 models." This change was disclosed last week, with Google's support documentation also referring to it as (3.5) Flash-Lite.

According to Google, Gemini will use auto model selection, sending each prompt to the model that best suits it. Most prompts will go to Flash-Lite, their fastest model, while prompts requiring deeper reasoning may be sent to either Flash or Pro. If smart model selection is disabled, chats for users without an AI plan will use Flash-Lite for all responses.

In addition to this change, Google is introducing new thinking levels that replace the previous "Extended" option. These three options align with Google Antigravity and AI Studio. Free users will now have access to these new thinking levels, which are also being rolled out to all subscribers.

However, Google AI Plus users will soon be limited to the same Flash-Lite model. It appears that the transition to these changes is happening rapidly, with the update now in place for the vast majority of free accounts in the Pacific Time zone. For more news on this development, viewers can check out 9to5Google on YouTube.

Written by urgent.news from 9to5Google's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at 9to5google.com →

More in AI

When Agent Chains Run for Hours: Why Checkpointing Is the Real Challenge

Today's GitHub Trending tells an interesting story. morluto/rea uses agent chains to reverse engineer anything "from app behavior down to native binaries." boykopovar/AnyPS5 automates PS5 executable…

  • Long-running agent workflows face significant challenges
  • Restarting from failure wastes hours of computation
  • Astron-agent provides checkpointing and resuming solutions

AI can design an app now. Are designers redundant?

AI can now draft a screen, write the button labels, and turn a sketch into a clickable prototype in minutes. So here is the question every designer has quietly asked at 2 a.m.: are designers now…

  • AI can rapidly generate design options, but lacks contextual understanding
  • AI excels at interface drafting but fails in problem definition, prioritization, and validation

How Content Creators Use ElevenLabs to Scale Video Production

Introduction If you’ve ever spent hours recording, editing, and re‑recording narration for a video, you know the pain of trying to keep a consistent tone, pacing, and energy level.

  • ElevenLabs API enables high-quality narration at scale for content creators.
  • AI voices eliminate need for professional voice actors and allow multilingual content.
  • Python example provided for integrating ElevenLabs into video production workflow.

More from Saturday 10 October →