meralion-merged-llama3-8B-sg-inst-avg-diff: benchmark results
Provider: A*STAR.
Aggregate Rating 1431 ± 38, rank #4608 of 10916 rated models, from 28 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SeaEval - Emotion - IndoEmotion (Zero-Shot) | 70 | Accuracy (%) | 66.7 |
| SeaEval - Dialogue - DialogSum (Zero-Shot) | 25.24 | Average ROUGE (0-100) | 65.2 |
| SeaEval - FLORES Translation - Indonesian-to-English (Zero-Shot) | 38.38 | BLEU (0-100) | 47.8 |
| SeaEval - Cultural Reasoning - US-Eval (Zero-Shot) | 73.83 | Accuracy (%) | 45.8 |
| SeaEval - Dialogue - SAMSum (Zero-Shot) | 28.28 | Average ROUGE (0-100) | 45.8 |
| SeaEval - Dialogue - DREAM (Zero-Shot) | 91.03 | Accuracy (%) | 43.5 |
| SeaEval - FLORES Translation - Chinese-to-English (Zero-Shot) | 24.13 | BLEU (0-100) | 43.5 |
| SeaEval - FLORES Translation - Malay-to-English (Zero-Shot) | 37.3 | BLEU (0-100) | 43.5 |
| SeaEval - FLORES Translation - Vietnamese-to-English (Zero-Shot) | 30.9 | BLEU (0-100) | 43.5 |
| SeaEval - Fundamental NLP Tasks - C3 (Zero-Shot) | 87.06 | Accuracy (%) | 37.5 |
| SeaEval - Multilingual Reasoning - ZBench (Zero-Shot) | 42.42 | Accuracy (%) | 35.4 |
| SeaEval - Cross-Lingual Consistency - Cross-XQuAD (Zero-Shot) | 93.7 | Accuracy (%) | 35 |
Interactive version: theaggregate.ai/model?slug=meralion-merged-llama3-8b-sg-inst-avg-diff · How It Works · Data refreshed daily, snapshot 2026-11-08.