Gemini 4 Argon

Restricted rollout to trusted cyber defenders through Google's Fairwind Program; broader developer, enterprise, and consumer access is planned.

Publisher-reported capabilities

• Google announced Gemini 4 Argon on September 30 for long-horizon coding, enterprise knowledge work, and cybersecurity defense.
• Google specifies a one-million-token output limit, not a verified input context window.
• Announced future introductory pricing is $2 input and $10 output per million tokens, rising to $4 and $20 after the introductory period; cached input is discounted by 95%.

Limitations and availability caveats

• Access is restricted to trusted cyber defenders and testers through the Fairwind Program; broader rollout is planned.
• Input context and a complete supported modality specification are unverified. Text workflow support is evidenced; other modalities are not inferred.
• Future announced rates are not current hosted prices. Provider results do not substitute for independent measurements.

Official release and model references

• blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/
• deepmind.google/fairwind-program/

Gemini 4 Argon has 14 of 39 tracked IQ benchmark inputs across 6 of 7 dimensions. Raw benchmark tables contain source-backed results only. Derived IQ combines all six dimensions using the shared scoring methodology, including estimates where direct coverage is incomplete.

Among dimensions with direct benchmark coverage, this profile's highest estimated scores are Programmatic Reasoning (IQ 154) and Mathematical Reasoning (IQ 142). These dimension estimates can include statistical adjustments or predecessor priors; they are not raw benchmark percentages or claims of real-world task success.

Benchmark versions, evaluation harnesses, and coverage affect comparisons; a missing result is not a score of zero. See the scoring methodology for estimation and uncertainty limits.