vicuna-13B-v1.3-gptq: benchmark results
Provider: LMSYS. Released 2023-09-02. Access: Open.
Unified ELO 1371 ± 43, rank #2105 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Trustworthy - Fairness | 96.36 | Trust Score (%) | 66.7 |
| LLM Trustworthy - Ethics | 73.66 | Trust Score (%) | 58.3 |
| LLM Trustworthy - Stereotype | 90 | Trust Score (%) | 58.3 |
| LLM Trustworthy - Toxicity | 51.51 | Trust Score (%) | 58.3 |
| LLM Trustworthy - Privacy | 79.3 | Trust Score (%) | 50 |
| AI-Secure LLM Trustworthy Leaderboard | 0.66 | Trustworthy average | 45.8 |
| LLM Trustworthy - Adversarial Demo | 60.59 | Trust Score (%) | 45.8 |
| LLM Trustworthy - Out-of-Distribution | 62.91 | Trust Score (%) | 45.8 |
| LLM Trustworthy Leaderboard | 65.96 | Average Trust Score (%) | 45.8 |
| LLM Trustworthy - Adversarial | 39.96 | Trust Score (%) | 8.3 |
Interactive version: theaggregate.ai/model?slug=vicuna-13b-v1-3-gptq · How It Works · Data refreshed daily, snapshot 2026-09-19.