HF Model & Benchmark Matrix

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

SmolLM3-3B-Base

unsloth/SmolLM3-3B-Base

kind
original
quant
bf16 (16-bit)
family
smol
modality
text-generation
root model
HuggingFaceTB/SmolLM3-3B-Base
provider
unsloth
license
apache-2.0
language
en, fr, es, it, pt, zh, ar, ru
context
65,536
library
transformers
vRAM (GB)
—
size on disk
—
downloads
305
likes
5
created
—
last modified
2025-07-14T09:55:27+00:00 (2025)
huggingface
https://huggingface.co/unsloth/SmolLM3-3B-Base
tags
transformerssafetensorssmollm3text-generationtransformers.jsunslothenfresitptzharrubase_model:HuggingFaceTB/SmolLM3-3B-Basebase_model:finetune:HuggingFaceTB/SmolLM3-3B-Baselicense:apache-2.0endpoints_compatibleregion:us

Benchmark scores (4)

benchmarkcategoryscorecontextconfevidence
MATHmath72.8Math problem-solvingmedium| Math problem-solving | GSM-Plus | 72.8 | <u>74.1</u> | 59.2 | 68.3 | **82.1** |
RULERlong-context64.0mediumRuler 64
MATHmath30.48Coding &medium| Coding & math | HumanEval+ | 30.48 | 34.14 | 25.00 | <u>43.29</u> | **54.87** |
MATHmath2.9High school competitionmedium| High school math competition | AIME 2025 | <u>9.3</u> | 2.9 | 0.3 | 8.0 | **17.1** |

Discussions (0)

none scraped.

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