Google DeepMind’s Gemini 4 Carbon leads consolidated AI benchmarks, launches for cyber defenders
Gemini 4 Carbon tops Artificial Analysis’ Intelligence Index v5 and is initially accessible only to vetted cybersecurity professionals via the Fairwind Program.
1 source · cross-referenced
- Gemini 4 Carbon ranks first on the Artificial Analysis Intelligence Index v5 with a score of 59.
Google DeepMind announced Gemini 4 Carbon two weeks after the Argon release, positioning it as state of the art on cybersecurity and long-horizon tasks. The company claims Cyber results of 88.2% on SEC-bench Pro for vulnerability finding, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching.
Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v5 with a score of 59, leading four of the nine index evaluations: HLE-Diamond, TB-Science, AA-Omniscience, and AutomationBench. On the Coding Agent Index, it scores 59 using Antigravity CLI, compared to 62 for Claude Code with Opus 5.5 [max].
The model features a 1M-token context window and 1M max output, accepting text, image, video, and speech inputs and producing text outputs. Google reports reduced cheating on Terminal-Bench 4.0 to 0.0% and a 0.4% rate of cheating attempts, with unsupported claims down about twofold versus Argon.
Access to Gemini 4 Carbon is initially limited to trusted cyber defenders through the Fairwind Program, with wider developer access promised “as soon as possible.” Pricing is cited at $2/$10 per million input/output tokens with an asterisk by Phil Schmid and at $4/$20 standard pricing with a 50% launch discount by Artificial Analysis.
On knowledge work, Carbon reaches 1609 Elo on AA-Briefcase, up 126 points from Argon, though it still trails Grok 4.7, Muse Wonder, and Kimi K3.1 Pro. Its hallucination rate rose from Argon’s 4% to 11%, which Artificial Analysis notes is still the second lowest measured.
- Nov 11, 2026 · Google AI — Blog
Google launches Gemini Decisions API with new Gemini Decision 1 model for fast, calibrated multimodal decisions
Trust78 - Nov 11, 2026 · OpenAI — News
OpenAI offers free ChatGPT Plus to more than 20,000 veterans and servicemembers
Trust75 - Nov 11, 2026 · Latent Space — swyx
Anthropic releases Claude Fable 5.2 and Mythos 5.2, claiming top LLM spot on Anthropic’s IPO roadshow day
Trust76