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

The post claims Gemini 4 Argon is the first model to lose LMArena votes weekly, dropping from 4932 to 4900 between Sep 29 and Oct 20, and alleges rankings follow Polymarket bets rather than votes.

iScienceLuvr says the two new frontier models, Gemini 4 Argon and Muse Wonder Full, tie for both first and last on CancerBench with a score of zero.

Command Code's weekly recap lists new VS Code and Cursor extensions, several Flash models and free stealth models, plus integration of Gemini 4 Argon and Meta's Muse Wonder.

A user frames Gemini 4 Argon and Muse Wonder Full as two new AI challengers and asks which has more potential.

The author shares a generated-video comparison of Gemini 4 Argon and Muse Wonder, joking that Muse Wonder should be free like Facebook.

Given the same prompt to build a 3D boss fight of a small knight against a clockwork colossus in the sky, Gemini 4 Argon and Muse Wonder produced results described as a studio trailer versus a prototype.

On the same prompt, Muse took about 40 minutes and cost around $30, while Gemini 4 Argon took about 30 minutes and cost around $10, and the author thinks Argon did better.

An Arabic-language creator gave the same prompt to Gemini 4 Argon and Muse Wonder to build a game each, and the results differ sharply, with a video comparison.

The poster shares a video comparing Gemini 4 Argon with Muse Wonder, with text only calling the results insane and no description of what was tested.

A joking post imagining how Gemini Argon would react when working on code generated by Meta's Muse Wonder.

The author criticizes paid hype in AI, says Muse Wonder Full felt strange to try, and argues Gemini 4 Argon was the week's real standout launch, calling for more builders to share honest tests.

The poster says a majority prefer Gemini 4 Argon over Meta's Muse Wonder, but expects everyone to switch to Meta once Muse Wonder-Full is available to all.