GPT-JT-6B-v1: benchmark results
Provider: EleutherAI. Access: Open.
Unified ELO 1320 ± 22, rank #2621 of 2928 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 47.19 | Accuracy (%) (5-shot) | 26.1 |
| Open LLM Leaderboard - MuSR | 37.37 | Score | 25.6 |
| Open LLM Leaderboard v1 - WinoGrande | 65.27 | Accuracy (%) (5-shot) | 18.5 |
| Open LLM Leaderboard v1 - HellaSwag | 67.15 | Normalized accuracy (%) (10-shot) | 18.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 40.87 | Normalized accuracy (%) (25-shot) | 17.2 |
| Open LLM Leaderboard - MMLU-Pro | 16.26 | Score | 16.9 |
| Open LLM Leaderboard - BBH | 33.03 | Score | 16 |
| Open LLM Leaderboard v1 - GSM8K | 1.21 | Accuracy (%) (5-shot) | 16 |
| Open LLM Leaderboard - GPQA | 26.09 | Score | 14.5 |
| Open LLM Leaderboard - IFEval | 20.61 | Score | 12.9 |
| Open LLM Leaderboard - MATH Level 5 | 1.06 | Score | 7.6 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 37.07 | MC2 (%) (0-shot) | 4.8 |
Interactive version: theaggregate.ai/model?slug=gpt-jt-6b-v1 · How It Works · Data refreshed daily, snapshot 2026-09-23.