Model

Falcon 7B

TII · open weights

falconllm.tii.ae

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BBH28.8averageN/AIndependentmediumarxiv.org
BoolQ75.3scoreN/AIndependentmediumcrfm.stanford.edu
BoolQ73.8scoreN/ASelf-reportedmediumarxiv.org
HellaSwag78.1overall_accuracyN/ASelf-reportedmediumarxiv.org
HellaSwag76.4overall_accuracyN/AIndependentmediumarxiv.org
HellaSwag76.3accuracyN/AIndependentmediumarxiv.org
LAMBADA74.9scoreN/ASelf-reportedmediumarxiv.org
MMLU35.0exact_matchN/ASelf-reportedmediumarxiv.org
MMLU28.0average_(%)N/AIndependentmediumarxiv.org
MMLU26.9exact_matchN/AIndependentmediumarxiv.org
OpenBookQA51.6accuracyN/AIndependentmediumarxiv.org
PIQA80.3scoreN/ASelf-reportedmediumarxiv.org
PIQA79.4scoreN/AIndependentmediumarxiv.org
Trivia QA64.6exact_matchN/AIndependentmediumarxiv.org
WinoGrande67.2accuracyN/AIndependentmediumarxiv.org
WinoGrande67.2accuracyN/ASelf-reportedmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K6.8exact_matchN/AIndependentmediumarxiv.org
GSM8K4.6exact_matchN/ASelf-reportedmediumarxiv.org

Reproduced by the community

Loading community runs…