HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Phi-3-small-8k-instruct

microsoft/Phi-3-small-8k-instruct

kind
original
quant
bf16 (16-bit)
family
phi
modality
text-generation
root model
microsoft/Phi-3-small-8k-instruct
provider
microsoft
license
mit
language
multilingual
context
8K
library
—
vRAM (GB)
—
size on disk
—
downloads
15,205
likes
178
created
—
last modified
2025-12-10T20:20:39+00:00 (2025)
huggingface
https://huggingface.co/microsoft/Phi-3-small-8k-instruct
tags
transformerssafetensorsphi3smalltext-generationnlpcodeconversationalcustom_codemultilinguallicense:mitregion:us

Benchmark scores (17)

benchmarkcategoryscorecontextconfevidence
ARC-Challengeknowledge90.7ARC Challenge<br>10-shotmedium| ARC Challenge<br>10-shot | 90.7 | 78.3 | 87.3 | 82.8 | 87.4 | 88.3 | 95.6 |
GSM8Kmath89.6GSM8K Chain of Thought<br>8-shotmedium| GSM8K Chain of Thought<br>8-shot | 89.6 | 59.8 | 64.7 | 77.4 | 78.1 | 80.4 | 94.2 |
OpenBookQAknowledge88.0OpenBookQA<br>10-shotmedium| OpenBookQA<br>10-shot | 88.0 | 78.6 | 85.8 | 82.6 | 86.0 | 86.4 | 93.4 |
PIQAknowledge86.9PIQA<br>5-shotmedium| PIQA<br>5-shot | 86.9 | 78.1 | 86.0 | 75.7 | 86.6 | 86.2 | 90.1 |
BoolQknowledge84.8BoolQ<br>2-shotmedium| BoolQ<br>2-shot | 84.8 | 66.0 | 76.6 | 80.9 | 79.1 | 86.4 | 91.3 |
WinoGrandeknowledge81.5WinoGrande<br>5-shotmedium| WinoGrande<br>5-shot | 81.5 | 55.6 | 62.0 | 65.0 | 68.8 | 72.2 | 86.7 |
CommonsenseQAknowledge80.0CommonsenseQA<br>10-shotmedium| CommonsenseQA<br>10-shot | 80.0 | 76.2 | 78.1 | 79.0 | 79.6 | 81.8 | 86.7 |
BBHreasoning79.1BigBench Hard<br>3-shotmedium| BigBench Hard<br>3-shot | 79.1 | 59.6 | 69.7 | 51.5 | 68.3 | 75.6 | 87.7 |
HellaSwagknowledge77.0HellaSwag<br>5-shotmedium| HellaSwag<br>5-shot | 77.0 | 49.8 | 70.4 | 71.1 | 78.8 | 76.2 | 88.3 |
MMLUknowledge75.7MMLU<br>5-shotmedium| MMLU<br>5-shot | 75.7 | 63.6 | 70.5 | 66.5 | 71.4 | 66.7 | 84.0 |
MBPPcoding71.7MBPP<br>3-shotmedium| MBPP<br>3-shot | 71.7 | 51.5 | 60.2 | 67.7 | 77.8 | 73.2 | 86.7 |
TruthfulQAsafety70.2TruthfulQA (MC2)<br>10-shotmedium| TruthfulQA (MC2)<br>10-shot | 70.2 | 52.1 | 60.1 | 63.2 | 67.7 | 72.6 | 85.2 |
MedQAdomain65.4MedQA<br>2-shotmedium| MedQA<br>2-shot | 65.4 | 49.6 | 62.2 | 60.5 | 63.4 | 58.2 | 83.7 |
HumanEvalcoding61.0HumanEval<br>0-shotmedium| HumanEval<br>0-shot | 61.0 | 34.1 | 37.8 | 60.4 | 62.2 | 64.4 | 79.9 |
TriviaQAknowledge58.1TriviaQA<br>5-shotmedium| TriviaQA<br>5-shot | 58.1 | 72.3 | 82.2 | 67.7 | 85.8 | 80.2 | 73.3 |
MATHmath51.6Mathmedium| Math | 51.6 | 35.8 | 40.3 | 41.1 | 52.8 | 50.9 | 67.1 |
AGIEvalreasoning45.1AGI Eval<br>5-shotmedium| AGI Eval<br>5-shot | 45.1 | 42.1 | 45.2 | 42.0 | 48.4 | 49.0 | 59.6 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.