From the source
- Google made Gemini 4 Carbon available to paying developers and enterprises and began rolling it out to Google AI Ultra subscribers, widening access that had been limited to a subset of cybersecurity defenders.
- The October 28, 2026 introduction ran through a deliberately limited Fairwind rollout, and the specialized cyber configuration stays gated behind Fairwind applications and rolling admission.
- The frontier model's introductory API price is $2 per million input tokens and $10 per million output tokens, including reasoning tokens, rising to $4 and $20 after November 30, 2026.
- Carbon is excluded from the Gemini API free tier, and Google's pricing page says every processed token is billed, including thinking tokens, so matching Argon's per-token rates does not guarantee matching per-task costs.
- The model card says Carbon builds on Argon's pre-training foundation with added post-training for deep reasoning and multi-step agent workflows, accepts text, images, audio and video, and lists hallucinations, possible slowness or timeouts and greater token consumption at higher thinking levels as limitations.
- Google's safety report says Carbon did not cross its material-capability-change threshold relative to Argon, and that strengthened safeguards reduce residual risk to an acceptable level, though it cannot be ruled out from reaching Cyber Uplift Level 1 without safeguards outside the model.