Tech giant Google recently quietly released its latest flagship AI model, Gemini 2.5 Pro, causing a significant stir within the developer community.
Google calls it "the most intelligent model ever," and this powerful model is now available in public preview via the Gemini API on Google AI Studio, with a soon-to-be-launched release on Vertex AI. This move not only allows more developers to experience top-tier AI capabilities but also sparks considerable industry interest due to its highly competitive pricing strategy.
While Google didn't detail all of Gemini 2.5 Pro's features in its announcement, being called "the most intelligent model ever" hints at its exceptional reasoning capabilities. The model shows promising potential in understanding complex problems, performing logical reasoning, and generating high-quality answers. Developers and users previously expressed strong impressions of its capabilities after experiencing the experimental version included in Gemini Advanced.
A Boon for Developers: Better Rates and Higher Efficiency
Responding to strong developer interest and early adoption, Google announced it would increase the rate limits for Gemini 2.5 Pro and offer the model at a price significantly lower than many competitors. Currently, the public preview version of Gemini 2.5 Pro is priced at $1.24 per million input tokens. Google previously mentioned that the experimental version would remain free, but with lower rate limits.
Compared to competitors, Gemini 2.5 Pro's pricing is remarkably attractive. For example, Anthropic's comparable model, Claude 3.7 Sonnet, costs $3 per million input tokens and $15 per million output tokens. OpenAI's o1 inference model is even higher, at $15 per million input tokens and $60 per million output tokens (with $7.5 for cached input).
Even OpenAI's more economical o3-mini model has higher input and output token prices than Gemini 2.5 Pro. This competitive pricing has led to social media users exclaiming, "Things are about to get interesting!"
AI Model Price War: Google Leads the Charge for Cost Reduction and Efficiency Improvement
Gemini 2.5 Pro's low-cost entry undoubtedly puts more pressure on other major AI model developers like Anthropic and OpenAI. Previously, DeepSeek caused a sensation in the enterprise world with its low-cost DeepSeek R1 model. Now, Google joins this "price war," aiming to make powerful AI models accessible to more developers at lower costs, thus promoting wider AI adoption.
Google states that by introducing Gemini 2.5 Pro into public preview, it aims to gather more developer feedback to further refine and optimize the model. This move also foreshadows even fiercer competition in the future of AI, with developers undoubtedly benefiting most from this technological advancement and price competition.
