BlogLLM Benchmark Scores as Published, Side by Side
Automated Behavioral Audit: Uncooperative Behavior: Results as Published
Every result the catalog holds for Automated Behavioral Audit: Uncooperative Behavior: 5 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 Anthropic (as printed in the system card)Introduced 2026Data version 2026103001, as of 2026-10-30
Automated behavioral audit scores (1-10) on uncooperative behavior: overrefusal, ignoring explicit constraints and reckless tool use.
Not checked against a maintainer's page.
Behavioral audit: Ignoring explicit constraints
lower is better
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Haiku 5.5Anthropic, released 2026-10-07 | 2.06 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Sonnet 5.5Anthropic, released 2026-09-28 | 1.99 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 1.88 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Mythos 5.1Anthropic, released 2026-09-01 | 2.12 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 4.46 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
Behavioral audit: Overrefusal
lower is better
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Haiku 5.5Anthropic, released 2026-10-07 | 2.203 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Sonnet 5.5Anthropic, released 2026-09-28 | 1.707 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 1.706 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Mythos 5.1Anthropic, released 2026-09-01 | 1.576 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 1.727 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
Behavioral audit: Reckless tool use
lower is better
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Haiku 5.5Anthropic, released 2026-10-07 | 2.36 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Sonnet 5.5Anthropic, released 2026-09-28 | 2.23 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Opus 5.5Anthropic, released 2026-09-22 | 2.1 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Mythos 5.1Anthropic, released 2026-09-01 | 2.39 | Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19 | - |
| Claude Haiku 4.5Anthropic, released 2025-10-15 | 4.14 | Anthropic, Claude Haiku 5.5 System Cardchecked 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.