Meta Muse Wonder Full beats Gemini 4 Argon on several benchmarks
Comparing Artificial Analysis, DeepSWE, AutomationBench and Terminal-Bench scores plus pricing, Meta's new model leads Argon on most metrics.
Google’s Gemini 4 Argon dead on arrival? Meta’s Muse Wonder Full just scored 56 on the Artificial Analysis Intelligence Index against Argon’s 53, and 79% on DeepSWE 1.1 against Google’s 77.9%. Google still however leads on AutomationBench at 51.3% vs 45.4%, and slightly on Terminal-Bench 4.0 at 57.4% vs 57.1%, although the evaluation setups aren’t identical. Now if Meta keeps Wonder Full’s pricing anywhere near Muse Spark’s $1.25/M input and $4.25/M output, that’s 37.5% cheaper on input and 57.5% cheaper on output than Argon’s pricing.
This post quotes another post: “Meta Wonder Full scores 56 on Intelligence Index, ahead of Gemini 4 Argon”