Google achieved the highest overall score of 59 on the index, propelled by top results in complex reasoning and scientific evaluations. Beyond these benchmarks, the system placed second on the Coding Agent Index and reached an index-leading score of 54 on AA-Omniscience v1.1. Operating at standard rates of $4 per million input tokens and $20 per million output tokens, it undercuts Claude Opus 5.5 by roughly 23% per task. The model is currently available to a limited group while Google provides temporary launch pricing.
Key points
Google's new model delivers elite multimodal processing capabilities across text, audio,…: Google's new model delivers elite multimodal processing capabilities across text, audio, video, and images alongside a massive 1-million-token context window.
The system achieves competitive pricing against Claude Opus 5.5 per task despite remainin…: The system achieves competitive pricing against Claude Opus 5.5 per task despite remaining more expensive than GPT-6 Astra.
Agentic knowledge capabilities improved substantially over Gemini 4 Argon by 126 Elo poin…: Agentic knowledge capabilities improved substantially over Gemini 4 Argon by 126 Elo points, even though it still trails rivals like Grok 4.7.
Sources checked by our analysis pipeline: 1
October 28, 2026
Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5
See model pageGoogle’s Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5, and second only to Claude Opus 5.5 on the Artificial Analysis Coding Agent Index
Gemini 4 Carbon scores 59 on Intelligence Index v5, driven by leading performance in complex reasoning, knowledge, and scientific tasks. It also has large gains in coding and knowledge work, particularly when compared to other Gemini models, but it is not leading in all these areas.
As with launch of Gemini 4 Argon, Gemini 4 Carbon is being rolled out to selected users and is not yet publicly available. Google is offering a 50% launch discount, with no confirmed end date. All our charts and figures use standard pricing for both Gemini 4 Carbon and Argon.
Key results for Gemini 4 Carbon with high reasoning:
➤ Gemini 4 Carbon leads the Intelligence Index v5, with the #1 score in 4 of the 9 Index evaluations: HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA. Carbon makes substantial…
Veteran content creator Seán McLoughlin is leveraging his massive digital audience to transition into traditional film production and interactive entertainment. By taking on character roles in major video games and guiding adaptations of beloved franchises, he aims to protect artistic authenticity while navigating a transforming entertainment landscape.
Faltering capital efficiency and emerging regulatory moats around frontier algorithms are prompting prominent market figures to curb risk across technology holdings. Mounting monetization challenges and deteriorating revenue-to-equipment ratios suggest the sector's rapid growth may soon stall. These fundamental strains threaten broader equity markets that have become deeply dependent on artificial intelligence momentum.
The French startup is taking on Chinese dominance in defensive artificial intelligence by offering an open-weights architecture that organizations can operate completely offline. This independent deployment model prevents external platforms from severing user access while helping corporate and state entities patch digital vulnerabilities directly. By releasing the weights publicly, the firm is ensuring unrestricted access while confronting the inherent security trade-offs of open distribution.