Gemini 4 Argon vs. Claude Opus 5.5: Half the price and about half as accurate
When Google announced Gemini 4 Argon to a small group of testers on September 30, it came with a benchmark The post Gemini 4 Argon vs. Claude Opus 5.5: Half the price and about half as accurate appeared first on The New Stack .
Google's Gemini 4 Argon model has emerged as a more affordable alternative to Anthropic's Claude Opus 5.5, according to a recent comparison. In a series of benchmark tests, Argon surpassed Opus 5.5 in 14 out of 18 tasks, such as DeepSWE and GraphWalks. However, Opus 5.5 outperformed Argon in four tests, including Terminal-bench 4.0 and Terminal-bench 4.0.
Google has also priced Argon at $2 per million input tokens and $10 per million output tokens, making it half as expensive as Opus 5.5. Despite this, Opus 5.5 remains superior in accuracy, particularly in the billing spec and long context tests, where it achieved a perfect score. For developers weighing the decision between the two models, the lower price of Argon may be appealing, but its accuracy discrepancies necessitate careful consideration.
Written by urgent.news from The New Stack's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
