Gemini 4 Carbon at a glance
Gemini 4 Carbon is Google's strongest model for long-horizon and cybersecurity work and tops Artificial Analysis' Intelligence Index, but access starts with a small group of cyber defenders.
Current standing
As of
- Rank
- #1of 43 models
Gemini 4 Carbon (High) · 60 on the Artificial Analysis Intelligence Index v5.0.1 (opens in a new tab)
- Recorded at publication
- 60 (v5.0.1)
as of
- Current leader
- Still the top model in this snapshot.
Computed from our daily snapshot; it changes as new models are measured. Google announces Gemini 4 Carbon, first for now to vetted cyber defenders
What changed with Gemini 4 Carbon
Google introduced Gemini 4 Carbon two weeks after Argon, using the same safety stack with extra training. Google says internal teams ran it for days on large codebases. It reported 88.2% on SEC-bench Pro, 54.8% on ExploitGym, 71.0% on CWE-bench and 0.0% attack success on Gray Swan's indirect prompt-injection test.
Access began with a subset of cyber defenders and will widen through the Fairwind Program. Google gave no date for developers, businesses or consumers. Announced pricing matches Argon: $2/$10 per million tokens at first, then $4/$20.
Gemini 4 Carbon benchmarks and pricing
Values are copied from the linked source on the date shown. They are not IA Compa measurements.
- Artificial Analysis Intelligence Indexas of · Source for Artificial Analysis Intelligence Index: artificialanalysis.ai (opens in a new tab)
- 60 (v5.0.1)
- Output speedas of · Source for Output speed: artificialanalysis.ai (opens in a new tab)
- 68 tokens/s
- Input price per 1M tokens (after intro period)as of · Source for Input price per 1M tokens (after intro period): blog.google (opens in a new tab)
- $4.00
- Output price per 1M tokens (after intro period)as of · Source for Output price per 1M tokens (after intro period): blog.google (opens in a new tab)
- $20.00
- Context windowas of · Source for Context window: artificialanalysis.ai (opens in a new tab)
- 1M tokens
How it compares
Other models in the same category that we have covered, using each model's latest recorded value. (Models)
Gemini 4 Carbon reasoning effort
Artificial Analysis tests this model at several reasoning-effort settings. Higher effort usually scores higher but costs more and answers more slowly.
Gemini 4 Carbon by effort level
Higher effort lets the model think longer. Here is what each setting scores, how fast it writes, and what it costs on Artificial Analysis.
| Effort | Intelligence Index | Output speed (tokens per second) | Cost to run the index | Blended price (3:1) (per 1M tokens) | Price in / out (per 1M tokens) | First token |
|---|---|---|---|---|---|---|
| High (Gemini 4 Carbon (high)) · Best: highest effort with a score, used in the leaderboard | 60 | 68 | $8,240 | $8 | $4 / $20 | 70 s |
| Medium (Gemini 4 Carbon (medium)) | 58 | 72 | $5,903 | $8 | $4 / $20 | 32 s |
| Low (Gemini 4 Carbon (low)) | 54 | 36 | $5,725 | $8 | $4 / $20 | 8.2 s |
Where to compare Gemini 4 Carbon
Independent leaderboards and official resources for comparing this model yourself. Rankings change often; each site shows its own current data.
Leaderboards
General leaderboards
Official resources
Why Gemini 4 Carbon matters for businesses
As of November 5, 2026, Artificial Analysis lists Carbon at #1 of 69 models with 60 on Intelligence Index v5.0.1, one point ahead of Claude Opus 5.5, at 68 tokens per second. Google's post cited 59 on v5.0. For now, businesses cannot build on Carbon; Argon and Opus 5.5 are the usable top-tier options.
Sources for this story
- Gemini 4 Carbon: Our next frontier model (opens in a new tab)Official
Google · blog.google
- Gemini 4 Carbon model page (opens in a new tab)Benchmark
Artificial Analysis · artificialanalysis.ai
- Artificial Analysis model leaderboard (opens in a new tab)Benchmark
Artificial Analysis · artificialanalysis.ai
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #llm
- #gemini
- #cybersecurity
- #frontier
