♊ Gemini 3.5 Flash Beats Google's Own Pro Model

The new Gemini 3.5 Flash is now Google's most powerful model, outperforming Gemini 3.1 Pro on most tasks.

On some benchmarks it even edges out Opus 4.7 and GPT-5.5, while running roughly 4x faster than flagship models from Anthropic and OpenAI.

💰 The API price has tripled though — up to $1.5/$9 per 1M input/output tokens. Still cheaper than the Pro tier, but no longer dirt cheap.

➡️ The model is free to use in Gemini and AI Studio, and it's now the default for Google's AI search mode.

Is Google leading the AI race again?

❤️ — Yes, this is a leap!
🔥 — Nah, they just stopped falling behind…