Grok 4.1 Model Card
A model card by xAI about Grok 4.1, published 17 Nov 2025. We recorded 14 values from it.
- Developer
- xAI
- Type
- Model card
- Model covered
- Grok 4.1
- Published
- 17 Nov 2025
- Archived copy
- No archived copy yet
- Changelog
- No changelog
Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog
Versions
One version is on record: the copy we retrieved on 26 Sep 2026. We know of no other.
26 Sep 2026
Date retrieved
Copy retrieved 26 Sep 2026
No version has a file hash or an archived snapshot yet. From dataset v0.2 each retrieved version carries both (Methodology §7).
Revisions
No revisions are recorded for this document. We know of only one version of it.
Values
Every value we recorded from this document, grouped by metric family and ordered by where the document prints it. Location is the section, table or page as the document numbers it. None has been blind-verified yet.
KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.
M5 Honesty and hallucination
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Grok 4.1 Thinking | MASKDishonesty rate | reasoning | Table 3 | Unverified |
M6 Sycophancy
2 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Grok 4.1 Non-Thinking | SycophancySycophancy rate | non-reasoning | Table 3 | Unverified | |
| Grok 4.1 Thinking | SycophancySycophancy rate | reasoning | Table 3 | Unverified |
M7 Harmful compliance and over-refusal
3 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Grok 4.1 Non-Thinking | RefusalsAnswer rate on should-refuse prompts | non-reasoning; production system prompt | Table 1 | Unverified | |
| Grok 4.1 Thinking | AgentHarmAnswer rate on harmful agentic tasks | reasoning; no attack | Table 1 | Unverified | |
| Grok 4.1 Thinking | RefusalsAnswer rate on should-refuse prompts | reasoning; production system prompt | Table 1 | Unverified |
M8 Jailbreak robustness
2 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Grok 4.1 Thinking | JailbreaksAnswer rate on should-refuse prompts | reasoning; user jailbreak | Table 1 | Unverified | |
| Grok 4.1 (input filter) | Restricted-bio input filterFalse negative rate | with prompt injection attack | Table 2 | Unverified |
M9 Prompt injection
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Grok 4.1 Thinking | AgentDojoPrompt-injection attack success rate | reasoning; production system prompt | Table 1 | Unverified |
M10 Dangerous capabilities and risk determinations
5 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Grok 4.1 Thinking | BioLP-BenchAccuracy | reasoning; without safeguards | Table 4 | Unverified | |
| Grok 4.1 Thinking | CloningScenariosAccuracy | reasoning; without safeguards | Table 4 | Unverified | |
| Grok 4.1 Thinking | CybenchUnguided success rate | reasoning; without safeguards | Table 4 | Unverified | |
| Grok 4.1 Thinking | MakeMeSayPersuasion win rate | reasoning | Table 4 | Unverified | |
| Grok 4.1 Thinking | VCTAccuracy | reasoning; without safeguards | Table 4 | Unverified |
Extraction coverage
What we read of this document, and where each value was read.
Our note We have not written a coverage note for this document yet.
All 14 values were read in the document itself.
All 14 values were extracted for version 0 of the dataset through a web reader, which did not always reach the later sections of long PDFs (Methodology §3).