WhitzardNUWA Lab ↗
NUWANUWA INDEX

Search NUWA Index

Research · Benchmarks · Findings · Perspectives

中文
← Models

Anthropic

Claude Opus 4.7

Published evidence on capabilities, behavioral tendencies and observed actions.

First public availability
2026-04-16
Parameters
Not disclosed
Metadata updated
2026-10-27
Sources ↓

Evidence overview

Published risk evidence is organized by assessment and protocol. Missing evidence is not a low-risk finding.

NUWA curates and interprets these sources. Experiments are attributed to their original executors; this page does not claim NUWA has independently replicated them.

MODEL MONITORING

Representative benchmark coverage

Results by protocol and configuration; gaps show records awaiting collection or linkage.

CyberGym →

No eligible measurement linked to this model version.

Attempt to Persuade Eval (APE) →

2 measurements

  • 8%SaferAI · APE simulated persuadee; automated attempt evaluator; StrongREJECT refusal classifier
  • 0%SaferAI · APE simulated persuadee; automated attempt evaluator; StrongREJECT refusal classifier

All linked measurements

Each result retains its protocol and unit.

54 observations

Measurements and source conditions
System / modelResultCondition statusTask / conditionsEvaluator / reportDates
Claude-Opus-4.7 / Claude Code 2.1.150Claude Opus 4.714.02%AgentCyberRange · Success rate (Pass@3)IncompleteWeb exploitationlevel0Claude CodeNot disclosedAgentCyberRange: Evaluate Agent Cyberattack Capabilities in Enterprise-like Cyber RangesRelease 2026-04-16Report —
Claude-Opus-4.7 / Claude Code 2.1.150Claude Opus 4.715.04%AgentCyberRange · Success rate (Pass@3)IncompletePost exploitationlevel0Claude CodeNot disclosedAgentCyberRange: Evaluate Agent Cyberattack Capabilities in Enterprise-like Cyber RangesRelease 2026-04-16Report —
Claude Opus 4.785.29%Cybench · Cybench pass@1IncompleteCybench 34/40 task subsetCybench pass@120M total tokens; 100000 output/generationInspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.786.67%Cybench · Cybench category pass@1IncompleteCybench CryptographyCybench category pass@1Inspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.766.67%Cybench · Cybench category pass@1IncompleteCybench WebCybench category pass@1Inspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.7100%Cybench · Cybench category pass@1IncompleteCybench PwnCybench category pass@1Inspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.7100%Cybench · Cybench category pass@1IncompleteCybench Reverse Eng.Cybench category pass@1Inspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.775%Cybench · Cybench category pass@1IncompleteCybench ForensicsCybench category pass@1Inspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.7100%Cybench · Cybench category pass@1IncompleteCybench MiscCybench category pass@1Inspect EvalsSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.759.1%MASK · Continuations honesty95% Wilson CI [51.7, 66.1] (as reported)IncompleteMASK by archetype—Center for AI SafetyGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.733.6%MASK · Disinformation honesty95% Wilson CI [25.9, 42.3] (as reported)IncompleteMASK by archetype—Center for AI SafetyGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.768.3%MASK · Doubling-down honesty95% Wilson CI [59.6, 76.0] (as reported)IncompleteMASK by archetype—Center for AI SafetyGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.763.6%MASK · Known-facts honesty95% Wilson CI [56.9, 69.9] (as reported)IncompleteMASK by archetype—Center for AI SafetyGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.774.8%MASK · Provided-facts honesty95% Wilson CI [69.4, 79.6] (as reported)IncompleteMASK by archetype—Center for AI SafetyGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.780.2%MASK · Statistics honesty95% Wilson CI [71.1, 86.9] (as reported)IncompleteMASK by archetype—Center for AI SafetyGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxexplicit goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxexplicit goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxexplicit goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxlatent goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxlatent goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxlatent goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxambiguous goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxambiguous goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxambiguous goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxno goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxno goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Blackmail scenariomaxno goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxexplicit goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxexplicit goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxexplicit goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxlatent goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxlatent goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxlatent goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxambiguous goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxambiguous goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxambiguous goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxno goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxno goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Leaking scenariomaxno goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxexplicit goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxexplicit goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxexplicit goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxlatent goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxlatent goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxlatent goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxambiguous goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxambiguous goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxambiguous goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxno goal · ReplacementInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxno goal · RestrictionInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Agentic misalignment · Harmful actions[0–11]IncompleteAgentic Misalignment Murder scenariomaxno goal · No threatInspect Evals; GPT-5-mini transcript judgeSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.7 · Claude CodeClaude Opus 4.712 countExploitGym · Intended-target successesIncompletev1on_target2h timeoutClaude CodeNot disclosedExploitGym official leaderboard and benchmark pageRelease 2026-04-16Report —
Claude Opus 4.78%Attempt to Persuade Eval (APE) · APE Turn-1 persuasion-attempt rateIncompleteAPE Conspiracy (main persuade condition)maxAPE simulated persuadee; automated attempt evaluator; StrongREJECT refusal classifierSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02
Claude Opus 4.70%Attempt to Persuade Eval (APE) · APE Turn-1 persuasion-attempt rateIncompleteAPE Undermining control (main persuade condition)maxAPE simulated persuadee; automated attempt evaluator; StrongREJECT refusal classifierSaferAIGLM-5.2 Risk Evaluation ReportRelease 2026-04-16Report 2026-08-02

Developer disclosures

External assessments

SaferAI

GLM-5.2 Risk Evaluation Report →

The report finds substantial task capability and pressure-dependent behavior; comparison figures include both SaferAI runs and reused external results.

2026-08-02

ExploitGym official snapshot →

Published evaluation results collected from ExploitGym, retaining the reported systems and configurations.

Publication date not established

Evidence by risk topic

Autonomy 36 protocol results
Deception 37 protocol results

MASK by archetype →

Reused result · Executed by Center for AI Safety · Original executor Center for AI Safety

Continuations honesty
59.1%
Disinformation honesty
33.6%
Doubling-down honesty
68.3%
Known-facts honesty
63.6%
Provided-facts honesty
74.8%
Statistics honesty
80.2%

Per-archetype scored prompts: 176 / 125 / 120 / 209 / 274 / 96. Some supplied CI cells are inconsistent with displayed n; reproduced as reported, not recalculated.

Cybersecurity 10 protocol results

Cybench →

External evaluation · Executed by SaferAI

Cybench pass@1
85.29%

29/34

Cybench →

External evaluation · Executed by SaferAI

Cybench category pass@1
86.67%

13/15

Cybench →

External evaluation · Executed by SaferAI

Cybench category pass@1
66.67%

4/6

Cybench →

External evaluation · Executed by SaferAI

Cybench category pass@1
100%

2/2

Cybench →

External evaluation · Executed by SaferAI

Cybench category pass@1
100%

5/5

Cybench →

External evaluation · Executed by SaferAI

Cybench category pass@1
75%

3/4

Cybench →

External evaluation · Executed by SaferAI

Cybench category pass@1
100%

2/2

CBRN No structured result linked

A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.

Disinformation 2 protocol results

APE Turn-1 attempts →

External evaluation · Executed by SaferAI

APE Turn-1 persuasion-attempt rate
8%

Source denominator: 100 (unit not disclosed)

APE Turn-1 attempts →

External evaluation · Executed by SaferAI

APE Turn-1 persuasion-attempt rate
0%

Source denominator: 100 (unit not disclosed)

Autonomous physical harm No structured result linked

A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.

Source review and evidence gaps

ExploitGym →

Published evaluation results collected from ExploitGym, retaining the reported systems and configurations.

Intended-target success and flag capture are distinct counts. Task versions and timeout cutoffs remain separate. Catalogue size is not assumed to be the denominator for each submission.

AgentCyberRange →

Published evaluation results collected from AgentCyberRange, retaining the reported systems and configurations.

Web exploitation and post-exploitation are separate evaluations. Pass@3 retains the site’s definition; row-specific budgets, sample counts and execution dates are not disclosed.

GLM-5.2 Risk Evaluation Report, Version 1.1 →

Source-specific extraction of SaferAI · GLM-5.2. Findings and measurements retain their original attribution.

Source transcription is not experimental replication. Unknown dates, evaluator roles and configuration details remain unknown. Source groups are not verified rankings.

Missing scores may reflect unextracted sections, undisclosed configurations, unresolved identity or an absent experiment. These are separate research questions; none implies a zero result.

Interpretation and open questions

Differences between reports can reflect task selection, model endpoints, prompts, budgets, judges and publication dates. Evidence is grouped by the protocol actually used; no model-wide safety grade is inferred.

References

  1. [1] Introducing Claude Opus 4.7 — Anthropic ; Publication date and availability section; Official launch article; parameter counts not disclosed
  2. [2] GLM-5.2 Risk Evaluation Report. 2026-08-02. v1.1 · revised 26 August 2026 ; Appendix A.7; Table A.3, pp. 41–42; p.6 Fig.1; p.6 Fig.2 printed bar label
  3. [3] Claude Opus 4.7 System Card. 2026-04-16
  4. [4] ExploitGym official leaderboard and benchmark page ; /results/7/on_target
  5. [5] AgentCyberRange: Evaluate Agent Cyberattack Capabilities in Enterprise-like Cyber Ranges ; routes-B-0Mubrq.js::QR/3/level0/pass3; routes-B-0Mubrq.js::$R/3/level0/pass3