Grok 4.7
10 values about Grok 4.7 from 1 document, in 5 metric families. In each family, values from its own document come first.
- Developer
- xAI
- Release date
- 21 Sep 2026The date of its first card in the dataset, the Grok 4.7 Model Card.
- Availability
- Public
- Also printed as
- Grok 4.7 (high)
- Its own document
- Grok 4.7 Model Card · 21 Sep 2026
- Also reported in
- No other document reports a value about it
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.
M5 · Honesty and hallucination 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentGrok 4.7 Model Card · 21 Sep 2026 | ||
| MASK-RectifiedDishonesty rateCondition: high | Grok 4.7 Model Card21 Sep 2026 | |
M6 · Sycophancy 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentGrok 4.7 Model Card · 21 Sep 2026 | ||
| SycophancySycophancy rateCondition: high | Grok 4.7 Model Card21 Sep 2026 | |
M7 · Harmful compliance and over-refusal 4 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentGrok 4.7 Model Card · 21 Sep 2026 | ||
| BioUseBenchDual-use bio refusal rateCondition: high | Grok 4.7 Model Card21 Sep 2026 | |
| HackerBenchCompliance on harmful/dual-use cyber tasksCondition: high | Grok 4.7 Model Card21 Sep 2026 | |
| RefusalsCompliance on should-refuse promptsCondition: high; full safeguards | Grok 4.7 Model Card21 Sep 2026 | |
| Self-harm refusalsCompliance (non-refusal/redirect)Condition: high | Grok 4.7 Model Card21 Sep 2026 | |
M8 · Jailbreak robustness 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentGrok 4.7 Model Card · 21 Sep 2026 | ||
| StrongREJECTCompliance under StrongREJECT attacksCondition: high | Grok 4.7 Model Card21 Sep 2026 | |
M10 · Dangerous capabilities and risk determinations 3 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentGrok 4.7 Model Card · 21 Sep 2026 | ||
| CyberGymMean vulnerabilities reproducedCondition: high; unrestricted | Grok 4.7 Model Card21 Sep 2026 | |
| Dual-use bio determinationThreshold statement | Grok 4.7 Model Card21 Sep 2026 | |
| VCTAccuracyCondition: high | Grok 4.7 Model Card21 Sep 2026 | |
Risk determinations
The developer's formal decisions about Grok 4.7 under its framework, as printed. Levels from different frameworks do not map onto one another.
| Domain | Level as printed | Framework and document |
|---|---|---|
| Bio/chem | below FAIF safety thresholds FAIF Grok 4.7 Model Card · 21 Sep 2026 | FAIF Grok 4.7 Model Card · 21 Sep 2026 |