Starcoder2 15B — Benchmarks

Benchmark scores for Starcoder2 15B aggregated from public leaderboards (4 boards), with how it ranks among open models and how it compares with its nearest neighbours. See hardware requirements for what you need to run it.

No llmrun Composite Score for Starcoder2 15B: the score is only published for a model measured on at least 4 of the boards in the score panel, spanning at least 2 benchmark categories and including a reasoning or coding board — or on 3 boards that narrow its score to a 90% range of at most 15.6 points. Either way the results must come from at least two different benchmarks. Its individual scores below stand on their own — see how the llmrun Composite Score gets calculated.

Starcoder2 15B reasoning benchmarks

Multi-step logic and problem-solving — can the model think through a hard problem rather than recall an answer.

WinoGrande
WinoGrande is a large-scale commonsense-reasoning test built around pronoun resolution: the model must pick which of two nouns a pronoun refers to in a sentence, using real-world knowledge to disambiguate. It was designed by AI2 to resist the shortcuts that let earlier models game its predecessor, Winograd.
64.3%
64.3%
median 70.1%open leader 89.2%
26th percentile · #38 of 51 open · #65 of 79 all
#38 of 51
#65 of 79
Lab-reported
Epoch AI · Feb 2024

Starcoder2 15B math benchmarks

Symbolic and quantitative problem-solving, from competition math to multi-step calculation.

GSM8K
GSM8K is a classic set of grade-school math word problems requiring several arithmetic reasoning steps. Long a default sanity check for reasoning, top models now score near the ceiling, so it best separates smaller models.
57.7%
57.7%
median 52.3%open leader 94.5%
57th percentile · #27 of 62 open · #36 of 93 all
#27 of 62
#36 of 93
Lab-reported
Epoch AI · Feb 2024

Starcoder2 15B knowledge benchmarks

Breadth and depth of factual knowledge across many subjects, paired with the reasoning to apply it.

MMLU
MMLU (Massive Multitask Language Understanding) spans 57 subjects from history to law to medicine as multiple-choice questions. It is the long-standing default for broad knowledge and remains the most widely-reported general benchmark.
64.1%
64.1%
median 62.5%open leader 87.2%
Overall best: GPT 4o (Nov 20, 2024) 88.1%
54th percentile · #40 of 85 open · #80 of 136 all
#40 of 85
#80 of 136
Lab-reported
Epoch AI · Feb 2024

Other results (1)· leaderboard-only, not part of the llmrun Score

ARC Challenge
ARC Challenge is a set of grade-school science questions, curated by the Allen Institute for AI (AI2) to exclude ones that simple word-matching or retrieval can answer. It measures science knowledge paired with the reasoning needed to apply it.
47.2%
47.2%
median 53.9%open leader 95.3%
40th percentile · #34 of 56 open · #51 of 76 all
#34 of 56
#51 of 76
Lab-reported
Epoch AI · Feb 2024

Scores aggregated from public benchmark sources (Epoch AI: 4 boards) — each linked from the benchmark pages. llmrun does not run these benchmarks. Percentiles and ranks are among open models in the llmrun catalogue; the median and leader marks refer to the same field.