Gemini 4 Pro / Flash: benchmarks, hands-on tests and leaks

The author compares Gemini 3.8 Flash and Gemini 4 Pro directly in Arena (listed under the 3.7 codename) and concludes that Gemini 4 Pro or Flash performs best.

A video compares outputs from Gemini 4 and Astra on the same prompt, and the author says the gap is getting hard to ignore.

A leak says Gemini 4 and the next Astra model will launch together next week, with Anthropic possibly shipping around the same time, seen as this summer's RSI results arriving at once.

A post says rumors have Gemini 4 Pro competing hard against Fable and Astra, and that Google may actually be making a real comeback this time.

A Korean user discovered that a Gemini subscription grants 1,000 Google Flow credits each month plus 50 per day, and also mentioned rumors that Gemini 4 Pro is very strong.

Shahbaztwt shared a screenshot of a Gemini 4 Pro benchmark, saying the first bar is not playing fair, without specifying which test it is.

A short reply to Google AI Studio says the user is waiting for Gemini 4 to be released before starting to build anything.

Quoting Google employee Tulsee Doshi: hype for Gemini 4 is building internally as teams start using the model.

A video says early reports show Gemini 4 Pro pushing hard on reasoning, coding, multimodal understanding and agentic tasks, which could reshape the top of the model race if it holds up in real-world testing.

One user speculates that some Arena results labeled gemini-3.7-flash are actually Gemini 4 Pro, but not every request routes there, which would explain why reported results swing between frontier-level and far behind.

A Japanese-language post says that after Gemini 3.8 went live, Gemini 4 Pro already seems available on Arena, and jokes that the holiday week will be busy.

A user says Gemini 4 Pro produced a beautiful mechanical butterfly from a three.js prompt, and notes it had been tested on Arena under the gemini-3.7-flash placeholder before apparently being removed.