HF Model & Benchmark Matrix

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

Phi-3-medium-4k-instruct

microsoft/Phi-3-medium-4k-instruct

kind
original
quant
bf16 (16-bit)
family
phi
modality
text-generation
root model
microsoft/Phi-3-medium-4k-instruct
provider
microsoft
license
mit
language
multilingual
context
4K
library
—
vRAM (GB)
—
size on disk
—
downloads
14,283
likes
226
created
—
last modified
2025-12-10T20:18:50+00:00 (2025)
huggingface
https://huggingface.co/microsoft/Phi-3-medium-4k-instruct
tags
transformerssafetensorsphi3text-generationnlpcodeconversationalcustom_codemultilinguallicense:miteval-resultstext-generation-inferenceendpoints_compatibleregion:us

Benchmark scores (17)

benchmarkcategoryscorecontextconfevidence
ARC-Challengeknowledge91.6ARC Challenge<br>10-shotmedium| ARC Challenge<br>10-shot | 91.6 | 86.9 | 91.3 | 93.0 | 87.4 | 88.3 | 95.6 |
GSM8Kmath91.0GSM8K Chain of Thought<br>8-shotmedium| GSM8K Chain of Thought<br>8-shot | 91.0 | 78.3 | 83.8 | 93.5 | 78.1 | 80.4 | 94.2 |
PIQAknowledge87.9PIQA<br>5-shotmedium| PIQA<br>5-shot | 87.9 | 86.4 | 85.0 | 85.3 | 86.6 | 86.2 | 90.1 |
OpenBookQAknowledge87.4OpenBookQA<br>10-shotmedium| OpenBookQA<br>10-shot | 87.4 | 86.8 | 88.6 | 91.8 | 86.0 | 86.4 | 93.4 |
BoolQknowledge86.5BoolQ<br>2-shotmedium| BoolQ<br>2-shot | 86.5 | 86.1 | 82.7 | 89.1 | 79.1 | 86.4 | 91.3 |
CommonsenseQAknowledge82.8CommonsenseQA<br>10-shotmedium| CommonsenseQA<br>10-shot | 82.8 | 82.0 | 82.0 | 84.4 | 79.6 | 81.8 | 86.7 |
HellaSwagknowledge82.4HellaSwag<br>5-shotmedium| HellaSwag<br>5-shot | 82.4 | 78.0 | 79.0 | 82.6 | 78.8 | 76.2 | 88.3 |
WinoGrandeknowledge81.5WinoGrande<br>5-shotmedium| WinoGrande<br>5-shot | 81.5 | 77.0 | 75.3 | 83.3 | 68.8 | 72.2 | 86.7 |
BBHreasoning81.4BigBench Hard<br>3-shotmedium| BigBench Hard<br>3-shot | 81.4 | 74.1 | 81.8 | 80.4 | 68.3 | 75.6 | 87.7 |
MMLUknowledge78.0MMLU<br>5-shotmedium| MMLU<br>5-shot | 78.0 | 73.8 | 76.2 | 80.2 | 71.4 | 66.7 | 84.0 |
MBPPcoding75.2MBPP<br>3-shotmedium| MBPP<br>3-shot | 75.2 | 68.9 | 70.7 | 81.3 | 77.8 | 73.2 | 86.7 |
TruthfulQAsafety75.1TruthfulQA (MC2)<br>10-shotmedium| TruthfulQA (MC2)<br>10-shot | 75.1 | 57.8 | 67.4 | 81.9 | 67.7 | 72.6 | 85.2 |
TriviaQAknowledge73.9TriviaQA<br>5-shotmedium| TriviaQA<br>5-shot | 73.9 | 82.8 | 84.5 | 78.5 | 85.8 | 80.2 | 73.3 |
MedQAdomain69.9MedQA<br>2-shotmedium| MedQA<br>2-shot | 69.9 | 59.2 | 67.9 | 78.5 | 63.4 | 58.2 | 83.7 |
HumanEvalcoding62.2HumanEval<br>0-shotmedium| HumanEval<br>0-shot | 62.2 | 61.6 | 39.6 | 78.7 | 62.2 | 64.4 | 79.9 |
MATHmath52.8Mathmedium| Math | 52.8 | 45.3 | 52.5 | 59.7 | 52.8 | 50.9 | 67.1 |
AGIEvalreasoning50.2AGI Eval<br>5-shotmedium| AGI Eval<br>5-shot | 50.2 | 50.1 | 54.0 | 56.9 | 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.