DeciLM-7B: benchmark results
Provider: Deci AI. Released 2023-12-10. Access: Open.
Unified ELO 1409 ± 1, rank #1101 of 1392 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open CoT - LSAT Analytical Reasoning | 6.52 | CoT Gain (%) | 81.3 |
| Open LLM Leaderboard - MuSR | 13.05 | Score | 71.7 |
| Open LLM Leaderboard - GPQA | 6.04 | Score | 50.7 |
| Open CoT - LSAT Reading Comprehension | 8.18 | CoT Gain (%) | 34.4 |
| Open LLM Leaderboard - BBH | 21.25 | Score | 31.7 |
| CyberMetric | 50.22 | Accuracy (%) | 29.2 |
| Open CoT - LSAT Logical Reasoning | 6.27 | CoT Gain (%) | 28.6 |
| Open LLM Leaderboard - MMLU-Pro | 18.8 | Score | 28.5 |
| Open CoT Leaderboard | 4.48 | Average CoT Gain (%) | 27.1 |
| Open LLM Leaderboard - IFEval | 28.13 | Score | 25.3 |
| Open CoT - LogiQA | 1.6 | CoT Gain (%) | 19.5 |
| Open LLM Leaderboard - MATH Level 5 | 2.87 | Score | 18.2 |
Interactive version: theaggregate.ai/model?slug=decilm-7b · How It Works · Data refreshed daily, snapshot 2026-09-05.