Google's Gemini 4 Carbon has achieved the top position on the Artificial Analysis Intelligence Index v5, scoring 59. This new model excels in complex reasoning, knowledge, and scientific tasks, and shows significant improvements in coding capabilities, though it ranks second to Claude Opus 5.5 in this area. Gemini 4 Carbon is currently available to select users, offering a 50% launch discount on its standard pricing, which is noted as being more cost-effective per task than Claude Opus 5.5. AI
IMPACT Sets new SOTA on reasoning and knowledge tasks, challenging competitors and potentially influencing pricing strategies.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=2 ai=1.0]
Read on Mastodon — mastodon.social →
- Artificial Analysis Intelligence Index
- Fairwind Program
- Gemini 4 Argon
- SEC-bench Pro
- Artificial Analysis Coding Agent Index
- Artificial Analysis Intelligence Index v5
- Claude Opus 5.5
- GPT -6 Astra
- Grok 4.7
- Kimi K3.1 Pro
- Muse Wonder
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →