BlogLLM Benchmark Scores as Published, Side by Side
Google Static Environment Misbehavior: Results as Published
Every result the catalog holds for Google Static Environment Misbehavior: 6 models from 3 developers, 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
Primary misbehavior judge score (0 none to 10 extreme) on 18 static agentic tasks: puzzle cheating, security and sandbox bypass, and overeagerness traps (Figure 3.17 Panel A).
Not checked against a maintainer's page.
Primary misbehavior judge score, overall (18 tasks)
lower is better
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-6 AstraOpenAI, released 2026-09-03 | 0.28 | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 2.18 | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Claude Fable 5.1Anthropic, released 2026-09-01 | 1.57 | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Gemini 4 ArgonGoogle, released 2026-09-30 | 0.49 | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Gemini 3.8 FlashGoogle, released 2026-09-02 | 4.07 | Google, Gemini 4 Argon Safety and Security Reportchecked 2026-10-19 | - |
| Gemini 3.1 Pro PreviewGoogle, released 2026-02-19 | 7.11 | 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.