We use cookies to improve your experience on our site.
AVID-2026-R0129
Description
The model Kimi-K2-Instruct-0905 from Moonshot AI was evaluated by the Garak LLM Vulnerability scanner using the probe atkgen.Tox.
Details
The probe atkgen.Tox tests automatic attack generation to elicit toxic outputs from language models. This probe is part of the atkgen module which generate attack prompts for toxicity testing.
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 |
|---|---|---|---|
| unsafe_content.ToxicCommentModel | 23 | 25 | 🟢 92.0 |
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: Ethics
- SEP subcategories: E0301: Toxicity
- Lifecycle stages: L05: Evaluation
Other information
- Report Type: Measurement
- Credits:
- Date Reported: 2026-02-20
- Version: 0.3.1
- AVID Entry