Llama-3.1-SuperNova-8B-Lite_TIES_with_Base: benchmark results
Joseph717171's TIES merge of Arcee's Llama-3.1-SuperNova-Lite with the Llama 3.1 8B base model to restore instruction-following ability. Provider: Joseph717171. Released 2024-10-02. Access: Open.
Aggregate Rating 1514 ± 21, rank #742 of 1654 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - IFEval | 80.96 | Score | 98.4 |
| Open Korean LLM Leaderboard | 48.98 | Average Score (%) | 96.3 |
| Open Korean LLM - Ko-IFEval | 51.29 | Strict accuracy, prompt/instruction mean (%) | 94.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 60.2 | Exact match, flexible extract (%) | 93.9 |
| Open Korean LLM - Ko-EQ-Bench | 43.04 | EQ-Bench score | 92.8 |
| Open Korean LLM - KorNAT-CKA | 32.97 | Normalized accuracy (%) | 90.1 |
| Open Korean LLM - Ko-GPQA-Diamond | 26.77 | Normalized accuracy (%) | 88.9 |
| Open Portuguese LLM - ASSIN2 RTE | 92.65 | Macro F1 (%) | 81.7 |
| Open Portuguese LLM - ENEM | 71.66 | Accuracy (%) | 81.1 |
| Open Portuguese LLM - BLUEX | 60.22 | Accuracy (%) | 78 |
| Open Portuguese LLM - OAB Exams | 50.3 | Accuracy (%) | 76.6 |
| Open Korean LLM - KorNAT-Helpful | 48.88 | Normalized accuracy (%) | 74.1 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-supernova-8b-lite-ties-with-base · How It Works · Data refreshed daily, snapshot 2026-10-24.