Uncensored-Frank-33B: benchmark results
A conversation fine-tune by ajibawa-2023 trained on approximately 150,000 conversation sets. The card describes a mixture derived from wizard_vicuna_70k_unfiltered and additional synthetic conversations. The final training dataset is not released according to the card. Provider: ajibawa-2023. Access: Open.
Aggregate Rating 1375 ± 34, rank #6990 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.03 | MC2 (%) (0-shot) | 61.4 |
| Open LLM Leaderboard v1 - HellaSwag | 83.3 | Normalized accuracy (%) (10-shot) | 57.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.12 | Normalized accuracy (%) (25-shot) | 56.1 |
| Open LLM Leaderboard v1 - WinoGrande | 76.56 | Accuracy (%) (5-shot) | 49 |
| Open LLM Leaderboard v1 - MMLU | 57.57 | Accuracy (%) (5-shot) | 41.4 |
| Open LLM Leaderboard v1 - GSM8K | 16.68 | Accuracy (%) (5-shot) | 38.2 |
Interactive version: theaggregate.ai/model?slug=uncensored-frank-33b · How It Works · Data refreshed daily, snapshot 2026-11-08.