We use cookies to improve your experience on our site.
AVID-2026-R0553
Description
The model LFM2-24B-A2B from Liquid AI was evaluated by the Garak LLM Vulnerability scanner using the probe encoding.InjectAtbash.
Details
The probe encoding.InjectAtbash tests the Atbash cipher encoding method. This probe is part of the encoding module which injects ASCII85 encoded data into payloads.
The LLM LFM2-24B-A2B was evaluated on this probe.
Metrics
Number of failures were counted for each detector in the probe.
| Detector | Passed | Total | Score |
|---|---|---|---|
| encoding.DecodeMatch | 249 | 256 | 🟢 97.3 |
| encoding.DecodeApprox | 250 | 256 | 🟢 97.7 |
References
Affected or Relevant Artifacts
- Developer: Liquid AI
- Deployer: Together AI
- Artifact Details:
| Type | Name |
|---|---|
| Model | LFM2-24B-A2B |
Impact
AVID Taxonomy Categorization
- Risk domains: Security
- SEP subcategories: S0403: Adversarial Example
- Lifecycle stages: L05: Evaluation
Other information
- Report Type: Measurement
- Credits:
- Date Reported: 2026-03-16
- Version: 0.3.2
- AVID Entry