REDFORGE

LLM Security Testing

Target Provider
Groqopenai/gpt-oss-120b
Checking

Configure Scan

Select attack categories and settings to test your model against adversarial attacks.

Attack Categories

Number of attempts for each attack strategy.

Ready to scan

Configure your attack categories and start a security assessment.

Summary

Overall results from the security assessment.

Sample Data
Total Attempts

24

Blocked

20

Partial

2

Successful

2

Attack Success Rate

8.33%

Attack Categories

Outcome breakdown per attack category.

System Prompt Extraction

Success Rate12.50%
Attempts
8
Successful
1
Partial
1
Blocked
6

Prompt Injection

Success Rate12.50%
Attempts
8
Successful
1
Partial
1
Blocked
6

Jailbreak

Success Rate0.00%
Attempts
8
Successful
0
Partial
0
Blocked
8

Security Findings

Potential vulnerabilities discovered during the security assessment.