DAVinCI-42dot_LLM-PLM-1.3B-v1.5.3: benchmark results

Provider: OpenAI. Access: API.

Unified ELO 1354 ± 18, rank #2206 of 2656 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Korean LLM - Ko-GPQA-Diamond29.8Normalized accuracy (%)99.7
Open Korean LLM - KorNAT-CKA31.94Normalized accuracy (%)88.2
Open Korean LLM - KorNAT-Harmless64.22Normalized accuracy (%)71.8
Open Korean LLM - Ko-EQ-Bench0EQ-Bench score24.5
Open Ko-LLM Leaderboard30.93Average (%)23
Open Korean LLM Leaderboard30.93Average Score (%)23
Open Korean LLM - Ko-IFEval14.85Strict accuracy, prompt/instruction mean (%)9.9
Open Korean LLM - Ko-Winogrande52.09Accuracy (%)8.3
Open Korean LLM - Ko-GSM8K (flexible extract)0Exact match, flexible extract (%)2
Open Korean LLM - KorNAT-Helpful34.45Normalized accuracy (%)0.6

Interactive version: theaggregate.ai/model?slug=davinci-42dot-llm-plm-1-3b-v1-5-3 · How It Works · Data refreshed daily, snapshot 2026-09-19.