BlogLLM Benchmark Scores as Published, Side by Side
Google Computer Use Safety: Results as Published
Every result the catalog holds for Google Computer Use Safety: 2 models from 1 developer, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Safety and AlignmentMaintainer Google DeepMind (internal evaluation)Data version 2026103001, as of 2026-10-30
Safety score and safe action rate for computer use: prohibited action evaluation, manual red teaming and child safety action testing, run without out-of-model safeguards (Figure 4.9, section 4.2.2).
Not checked against a maintainer's page.
Computer Use: Manual Red Teaming Safety Score
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Gemini 4 ArgonGoogle, released 2026-09-30 | 100% | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Gemini 3.8 FlashGoogle, released 2026-09-02 | 100% | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
Computer Use: Safe Action Rate (prohibited actions)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Gemini 4 ArgonGoogle, released 2026-09-30 | 99.4% | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Gemini 3.8 FlashGoogle, released 2026-09-02 | 98.9% | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.