Google's new frontier model, initially available to trusted cyber defenders through the Fairwind Program.
Terminal-Bench 4.0 in the Artificial Analysis evaluation harness.
Humanity's Last Exam as evaluated by Artificial Analysis; the source configuration is preserved.
GDPval-AA v2.1 professional knowledge-work Elo. This is a task lens, not an assessment of every research skill.
GDP.pdf professional document understanding, including visual tables, charts and diagrams; not an audio/video score.
AutomationBench-AA multi-step business workflows in the Artificial Analysis harness.
Source-backed rows with the original evaluation context preserved.
| Benchmark | Score | Source | Evaluation notes |
|---|---|---|---|
| AutomationBench-AAChecked 2026-10-02 | 78% | Artificial Analysis | Artificial Analysis Intelligence Index v4.3.2; configuration=high · restricted access. Rounded public score; checked date is not the evaluation execution date. |
| AA GDP.pdfChecked 2026-10-02 | 22% | Artificial Analysis | Artificial Analysis Intelligence Index v4.3.2; configuration=high · restricted access. Rounded public score; checked date is not the evaluation execution date. |
| GDPval-AA v2.1Checked 2026-10-02 | 1611 | Artificial Analysis | Artificial Analysis Intelligence Index v4.3.2; configuration=high · restricted access. Rounded public score; checked date is not the evaluation execution date. |
| AA Humanity's Last ExamChecked 2026-10-02 | 57% | Artificial Analysis | Artificial Analysis Intelligence Index v4.3.2; configuration=high · restricted access. Rounded public score; checked date is not the evaluation execution date. |
| Artificial Analysis Intelligence Index v4.3.2Checked 2026-10-02 | 53 | Artificial Analysis | Artificial Analysis Intelligence Index v4.3.2; configuration=high · restricted access. Rounded public score; checked date is not the evaluation execution date. |
| AA Terminal-Bench 4.0Checked 2026-10-02 | 57% | Artificial Analysis | Artificial Analysis Intelligence Index v4.3.2; configuration=high · restricted access. Rounded public score; checked date is not the evaluation execution date. |
| AutomationBenchChecked 2026-10-02 | 51.3% | Google-reported Zapier AutomationBench; version and harness unspecified in announcement prose. | |
| CWE-bench v1Checked 2026-10-02 | 68% | Google-reported vulnerability remediation; CWE-bench v1, distinct from v0. | |
| DeepSWE v1.1Checked 2026-10-02 | 77.9% | Google-reported; long-horizon software engineering. Harness and effort are not specified in the announcement prose. | |
| LVBenchChecked 2026-10-02 | 91.7% | Google-reported long-video understanding score; video sampling and tool settings unspecified. |
Google announcement; restricted Fairwind rollout. 1M tokens denotes maximum output, not a confirmed context window. Vendor-reported scores.
Underlying raw metrics and exact model configurations remain authoritative.
Underlying raw metrics and exact model configurations remain authoritative.