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

A Gemini 4 Carbon checkpoint spotted on Arena under the name "Gemini 4 Argon" renders an SVG of a pelican riding a bike through four seasons, shown in a video.

Wiz found two CVSS 9.2 PHP remote-code flaws affecting PHP 5.4+, using Gemini 4 Argon and GPT-5.6 Sol to help discover, test and verify fixes.

A reply to steipete says the team behind a discovery credits Gemini 4 Argon with doing the lion's share of the work amid others claiming credit.

The author splits a static city image into background, midground, and foreground depth plates to create a parallax animation, crediting Gemini 4 Argon for the technique and demo.

A roundup lists 10 examples of people building 3D worlds, robots and games with Gemini 4 Argon, framing it as Google's comeback.

Pichai said on the Big Technology Podcast that Gemini 4 Argon consumed 25 trillion tokens in its first 72 hours, about 3x the prior generation, and admitted compute is short every day; the post links a 3m40s Opus-made summary.

Opus 5.5, Gemini 4 Argon, DeepSeek V4.1 and MiniMax M3.1 each built a jet dogfight video from one prompt; Argon cost $2 and took 21 minutes.

A user recounts playing 20 Questions with Gemini 4 Argon, which nailed the answer (a floppy disk) by question 10 and then gloated about it.

Google says Gemini 4 Argon helped identify and mitigate critical security vulnerabilities in PHP, highlighting AI's role in software security.

A user says Gemini 4 Carbon is already on Arena with noticeable speed and quality improvements, hoping Google ships upgrades every 1–2 months.

Sundar Pichai says Gemini 4 Argon processed nearly 25 trillion tokens in its first 72 hours, about 3x Gemini 3, while Google Cloud forgoes short-term revenue to supply DeepMind.

A short image post where the author can't decide between Gemini 4 Argon and GPT Astra, offering no hands-on results or new details.