gpt2-chatbot: benchmark results
Provider: OpenAI. Access: API.
Unified ELO 1219 ± 20, rank #2870 of 2928 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BenchTable | 77.1 | Total Score (%) | 94.8 |
| BenchTable - Reasoning | 75.6 | Weighted Score (%) | 90.6 |
| BenchTable - STEM | 73.3 | Weighted Score (%) | 82.6 |
| BenchTable - Tech | 77.4 | Weighted Score (%) | 82.6 |
| BenchTable - Utility | 66.9 | Weighted Score (%) | 76 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 40.38 | MC2 (%) (0-shot) | 13.9 |
| Open LLM Leaderboard v1 - MMLU | 26.42 | Accuracy (%) (5-shot) | 12.5 |
| Open LLM Leaderboard v1 - HellaSwag | 50.27 | Normalized accuracy (%) (10-shot) | 10.4 |
| Open LLM Leaderboard v1 - WinoGrande | 56.67 | Accuracy (%) (5-shot) | 10 |
| Open LLM Leaderboard v1 - ARC Challenge | 29.69 | Normalized accuracy (%) (25-shot) | 9 |
| Open LLM Leaderboard v1 - GSM8K | 0 | Accuracy (%) (5-shot) | 4.6 |
Interactive version: theaggregate.ai/model?slug=gpt2-chatbot · How It Works · Data refreshed daily, snapshot 2026-09-23.