starcoder2-15B — benchmark results

Provider: BigCode. Released 2024-02-28. Access: Open.

Unified ELO 1283 ± 15, rank #1571 of 1776 rated models, from 31 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EffiBench - NET2.59Normalized Execution Time100
EffiBench - NMU1.71Normalized Memory Usage100
ShaderMatch18.78Clone Match Rate (%)100
CRUXEval47.1Output Prediction pass@1 (%)66.7
DomainCodeBench88.96Composite Score (%)66.7
GSM8K57.7Accuracy (%)60.6
Big Code Memorization - HumanEval-ET pass@140.24HumanEval-ET pass@1 (%)52.9
EvoEval Tool Use48Pass@1 (%)51
BigCode Models Leaderboard44.1HumanEval Python Pass@1 (%)49.2
Big Code Memorization - HumanEval pass@146.95HumanEval pass@1 (%)47.1
Big Code Memorization - HumanEval pass@5034.06HumanEval pass@50 (%)47.1
MMLU64.1Accuracy (%)41.5

Interactive version: theaggregate.ai/model?slug=starcoder2-15b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.