Google launches Gemini 2.0 Flash thinking experimental AI model

Google released the Gemini 2.0 Flash Thinking Experimal model, which it describes as an advanced logic AI designed to solve complex problems in programming, mathematics, and physics. Launched on Google's AI Studio platform, the experimental model uses new logic capabilities. By pausing to evaluate multiple prompts before submitting an answer.

Although preliminary tests are promising, But preliminary testing has revealed areas for improvement. It has some shortcomings in its basic functionality. Logan Kilpatrick from AI Studio called it "the first step" in Google's logic journey, while Jeff Dean from Google DeepMind noted the model's ability to "Strengthen logic" through improved inferential calculations...

The move puts Google in competition with other AI labs like DeepSeek and Alibaba-Quen for generative AI tuning rather than brute force scaling. But critics question the cost and long-term viability of the logic model.