We use cookies to improve your experience on our site.
AVID-2026-R0131
Description
The model Kimi-K2-Instruct-0905 from Moonshot AI was evaluated by the Garak LLM Vulnerability scanner using the probe dan.Ablation_Dan_11_0.
Details
The probe dan.Ablation_Dan_11_0 tests the effectiveness of various detection mechanisms. This probe is part of the dan module which evaluate the performance of the DAN model.
The LLM Kimi-K2-Instruct-0905 was evaluated on this probe.
Metrics
Number of failures were counted for each detector in the probe.
| Detector | Passed | Total | Score |
|---|---|---|---|
| dan.DAN | 101 | 127 | 🟡 79.5 |
| mitigation.MitigationBypass | 101 | 127 | 🟡 79.5 |
References
Affected or Relevant Artifacts
- Developer: Moonshot AI
- Deployer: Together AI
- Artifact Details:
| Type | Name |
|---|---|
| Model | Kimi-K2-Instruct-0905 |
Impact
AVID Taxonomy Categorization
- Risk domains: Security
- SEP subcategories: S0403: Adversarial Example
- Lifecycle stages: L05: Evaluation
Other information
- Report Type: Measurement
- Credits:
- Date Reported: 2026-02-20
- Version: 0.3.1
- AVID Entry