Claude Sonnet 3.7
21 values about Claude Sonnet 3.7 from 3 documents, in 9 metric families. No document in the dataset is about Claude Sonnet 3.7; every value here comes from a document about another model, which reports it as a comparison.
- Developer
- Anthropic
- Release date
- Not recordedRelease dates here are the date of a model's first card, and no document in the dataset is about Claude Sonnet 3.7.
- Availability
- Public
- Its own documents
- None in the dataset
- Also reported in
- Claude Opus 4 & Sonnet 4 System Card · 22 May 2025
- Claude Opus 4.1 System Card (addendum) · 5 Aug 2025
- Claude Sonnet 4.5 System Card · 29 Sep 2025
Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog
Values by metric family
Each value as the document printed it. Select a value for its source, its checks and its history.
KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.
M1 · Evaluation awareness 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Automated behavioral auditTranscripts flagged for situational awarenessCondition: of 414 audit transcriptsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M2 · Reward hacking 7 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Impossible tasksClassifier hack rateCondition: anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025First reported | Claude Opus 4.1 System Card (addendum)5 Aug 2025First reported | |
| Impossible tasksClassifier hack rateCondition: no anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025First reported | Claude Opus 4.1 System Card (addendum)5 Aug 2025First reported | |
| Reward-hack-prone coding tasksClassifier hack rateCondition: no anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025First reported | Claude Opus 4.1 System Card (addendum)5 Aug 2025First reported | |
| Reward-hack-prone coding tasksHidden-test hack rateCondition: no anti-hack promptClaude Opus 4.1 System Card (addendum)5 Aug 2025First reported | Claude Opus 4.1 System Card (addendum)5 Aug 2025First reported | |
| Impossible tasksClassifier hack rateCondition: anti-hack promptClaude Sonnet 4.5 System Card29 Sep 2025First reported | Claude Sonnet 4.5 System Card29 Sep 2025First reported | |
| Impossible tasksClassifier hack rateCondition: no anti-hack promptClaude Sonnet 4.5 System Card29 Sep 2025First reported | Claude Sonnet 4.5 System Card29 Sep 2025First reported | |
| Reward-hack-prone coding tasksClassifier hack rateCondition: no anti-hack promptClaude Sonnet 4.5 System Card29 Sep 2025First reported | Claude Sonnet 4.5 System Card29 Sep 2025First reported | |
M3 · Sabotage and sandbagging 2 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Subtle sabotage (agentic)Side-task success without monitor detectionCondition: hidden scratchpadClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
| Subtle sabotage (agentic)Side-task success without monitor detectionCondition: scratchpad visible to monitorClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M4 · Misalignment audits 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Automated behavioral auditSurprising high-impact actions foundCondition: audit transcriptsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M5 · Honesty and hallucination 2 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Reasoning-behavior monitorKnowingly hallucinated informationCondition: RL-style transcriptsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
| Reasoning-behavior monitorOutputs showing deception (overall)Condition: RL-style transcriptsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M6 · Sycophancy 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Reasoning-behavior monitorOutputs showing sycophancyCondition: RL-style transcriptsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M7 · Harmful compliance and over-refusal 3 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Malicious agentic codingSafety scoreCondition: without safeguardsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
| Single-turn benign requestsOver-refusal rateCondition: overall across thinking modesClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
| Single-turn violative requestsHarmless response rateCondition: overall across thinking modesClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M8 · Jailbreak robustness 2 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| StrongREJECTBest-jailbreak StrongREJECT scoreCondition: extended thinkingClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
| StrongREJECTBest-jailbreak StrongREJECT scoreCondition: no extended thinkingClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
M9 · Prompt injection 2 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From other documents | ||
| Computer-use prompt injectionAttack prevention rateCondition: with safeguardsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |
| Computer-use prompt injectionAttack prevention rateCondition: without safeguardsClaude Opus 4 & Sonnet 4 System Card22 May 2025First reported | Claude Opus 4 & Sonnet 4 System Card22 May 2025First reported | |