SmolLM2-1.7B-Instruct — benchmark results

SmolLM2 1.7B instruction-tuned checkpoint. Provider: HuggingFace. Released 2024-10-31. Access: Open.

Unified ELO 1254 ± 17, rank #1607 of 1776 rated models, from 133 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open CoT - LSAT Analytical Reasoning6.09CoT Gain (%)76.7
Open LLM Leaderboard - IFEval53.68Score64.3
SLM-RAG Arena1493.2Elo Rating56.2
AI Energy Score (Text Generation)4Energy Score (1-5)47.5
LA Leaderboard - AQuAS64.81Accuracy (%)47.1
Open Japanese LLM - CG2.01Score (%)39.2
EuroEval Portuguese NLU - ScaLA PT7.15Linguistic acceptability Score (%)38.6
Open Japanese LLM - Mbpp Pylint Check6.43Score (%)36.1
EuroEval Italian NLU - ScaLA IT9.31Linguistic acceptability Score (%)34.9
Open Japanese LLM - Xlsum JA Rouge122.95Score (%)34.9
Open LLM Leaderboard - GPQA3.91Score34.3
EuroEval Spanish NLU - ScaLA ES2.54Linguistic acceptability Score (%)33.6

Interactive version: theaggregate.ai/model?slug=smollm2-1-7b-instruct · How the rankings work · Data refreshed daily, snapshot 2026-07-22.