Housing-Subscription-QA-Phi-3.5: benchmark results
Provider: Other. Access: Open.
Unified ELO 1301 ± 17, rank #2458 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 21.8 | Normalized accuracy (%) | 36.6 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
| Open Korean LLM - KorNAT-Harmless | 60.94 | Normalized accuracy (%) | 21.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 1.59 | Exact match, flexible extract (%) | 14.5 |
| Open Korean LLM - Ko-IFEval | 16.05 | Strict accuracy, prompt/instruction mean (%) | 11.1 |
| Open Ko-LLM Leaderboard | 28.35 | Average (%) | 6.1 |
| Open Korean LLM Leaderboard | 28.35 | Average Score (%) | 6.1 |
| Open Korean LLM - KorNAT-Helpful | 39.62 | Normalized accuracy (%) | 5.4 |
| Open Korean LLM - Ko-Winogrande | 50.28 | Accuracy (%) | 3.3 |
Interactive version: theaggregate.ai/model?slug=housing-subscription-qa-phi-3-5 · How It Works · Data refreshed daily, snapshot 2026-09-19.