SmolLM2-360M-Instruct-GGUF

Home/Models/Mungert/SmolLM2-360M-Instruct-GGUF

Mungert/SmolLM2-360M-Instruct-GGUF

kind: gguf
quant: gguf (16-bit)
family: smol
modality: text-generation
root model: HuggingFaceTB/SmolLM2-360M
provider: Mungert
license: apache-2.0
language: en
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 150
likes: 0
created: —
last modified: 2025-09-24T15:43:01+00:00 (2025)
huggingface: https://huggingface.co/Mungert/SmolLM2-360M-Instruct-GGUF
tags: transformersggufsafetensorsonnxtransformers.jstext-generationenarxiv:2502.02737base_model:HuggingFaceTB/SmolLM2-360Mbase_model:quantized:HuggingFaceTB/SmolLM2-360Mlicense:apache-2.0endpoints_compatibleregion:usconversational

benchmark	category	score	context	conf	evidence
PIQA	knowledge	71.7	PIQA	medium	`\| PIQA \| 71.7 \| 69.9 \| 71.6 \|`
HellaSwag	knowledge	54.5	HellaSwag	medium	`\| HellaSwag \| 54.5 \| 51.2 \| 51.8 \|`
WinoGrande	knowledge	52.5	Winogrande	medium	`\| Winogrande \| 52.5 \| 54.1 \| 52.8 \|`
IFEval	instruction-following	41.0	IFEval (Average prompt/inst)	medium	`\| IFEval (Average prompt/inst) \| 41.0 \| 31.6 \| 19.8 \|`
CommonsenseQA	knowledge	38.0	CommonsenseQA	medium	`\| CommonsenseQA \| 38.0 \| 31.6 \| 35.3 \|`
OpenBookQA	knowledge	37.4	OpenBookQA	medium	`\| OpenBookQA \| 37.4 \| 37.4 \| 37.2 \|`
MMLU	knowledge	35.8	MMLU (cloze)	medium	`\| MMLU (cloze) \| 35.8 \| 33.7 \| 34.4 \|`
BBH	reasoning	27.3	BBH (3-shot)	medium	`\| BBH (3-shot) \| 27.3 \| 30.7 \| 24.4 \|`
TriviaQA	knowledge	16.9	TriviaQA	medium	`\| TriviaQA \| 16.9 \| 4.3 \| 9.1 \|`
MT-Bench	chat	3.66	MT-Bench	medium	`\| MT-Bench \| 3.66 \| 4.16 \| 3.37 \|`
GSM8K	math	3.2	GSM8K (5-shot)	medium	`\| GSM8K (5-shot) \| 3.2 \| 33.4 \| 1.6 \|`

none scraped.