t5gemma-2-1b-1b

Home/Models/google/t5gemma-2-1b-1b

t5gemma-2-1b-1b

google/t5gemma-2-1b-1b

kind: original
quant: bf16 (16-bit)
family: gemma
modality: image-text-to-text
root model: google/t5gemma-2-1b-1b
provider: google
license: gemma
language: —
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 25,388
likes: 81
created: —
last modified: 2025-12-18T18:12:04+00:00 (2025)
huggingface: https://huggingface.co/google/t5gemma-2-1b-1b
tags: transformerssafetensorst5gemma2text2text-generationimage-text-to-textarxiv:1905.07830arxiv:1905.10044arxiv:1911.11641arxiv:1904.09728arxiv:1705.03551arxiv:1911.01547arxiv:1907.10641arxiv:1903.00161arxiv:2009.03300arxiv:2304.06364arxiv:2103.03874arxiv:2110.14168arxiv:2311.12022arxiv:2108.07732arxiv:2107.03374arxiv:2210.03057arxiv:2502.12404arxiv:2106.03193arxiv:1910.11856

Benchmark scores (22)

benchmark	category	score	context	conf	evidence
PIQA	knowledge	66.5	[PIQA]	medium	`\| [PIQA] \| 0-shot \| 66.5 \| 74.3 \| 79.2 \|`
BoolQ	knowledge	57.4	[BoolQ]	medium	`\| [BoolQ] \| 0-shot \| 57.4 \| 68.5 \| 79.3 \|`
RULER	long-context	57.3	[Ruler 32K]	medium	`\| [Ruler 32K] \| 57.3 \| 69.2 \| 81.7 \|`
WinoGrande	knowledge	53.9	[WinoGrande]	medium	`\| [WinoGrande] \| 5-shot \| 53.9 \| 60.8 \| 71.6 \|`
DocVQA	multimodal	41.6	[DocVQA] (val)	medium	`\| [DocVQA] (val) \| 41.6 \| 66.6 \| 74.7 \|`
HellaSwag	knowledge	41.1	[HellaSwag]	medium	`\| [HellaSwag] \| 10-shot \| 41.1 \| 62.9 \| 77.4 \|`
DROP	reasoning	39.4	[DROP]	medium	`\| [DROP] \| 1-shot \| 39.4 \| 51.2 \| 66.7 \|`
RULER	long-context	32.0		medium	`Ruler 32`
ChartQA	multimodal	29.2	[ChartQA]	medium	`\| [ChartQA] \| 29.2 \| 50.2 \| 66.0 \|`
RULER	long-context	25.5	[Ruler 128K]	medium	`\| [Ruler 128K] \| 25.5 \| 35.1 \| 57.6 \|`
MMLU	knowledge	23.5	[Global-MMLU-Lite]	medium	`\| [Global-MMLU-Lite] \| 23.5 \| 33.1 \| 53.5 \|`
BBH	reasoning	22.9	[BIG-Bench Hard]	medium	`\| [BIG-Bench Hard] \| few-shot \| 22.9 \| 28.5 \| 43.7 \|`
MMMU	multimodal	22.7	[MMMU] (pt)	medium	`\| [MMMU] (pt) \| 22.7 \| 28.4 \| 39.4 \|`
AGIEval	reasoning	20.9	[AGIEval]	medium	`\| [AGIEval] \| 3-5-shot \| 20.9 \| 23.8 \| 41.1 \|`
TriviaQA	knowledge	14.8	[TriviaQA]	medium	`\| [TriviaQA] \| 5-shot \| 14.8 \| 29.0 \| 53.1 \|`
MMLU	knowledge	10.6	[MMLU] (Pro CoT)	medium	`\| [MMLU] (Pro CoT) \| 5-shot \| 10.6 \| 16.1 \| 33.2 \|`
GPQA	reasoning	9.6	[GPQA]	medium	`\| [GPQA] \| 5-shot \| 9.6 \| 10.5 \| 17.6 \|`
MBPP	coding	5.6	[MBPP]	medium	`\| [MBPP] \| 3-shot \| 5.6 \| 25.2 \| 44.8 \|`
HumanEval	coding	4.3	[HumanEval]	medium	`\| [HumanEval] \| 0-shot \| 4.3 \| 15.2 \| 30.5 \|`
MGSM	math	1.8	[MGSM]	medium	`\| [MGSM] \| 1.8 \| 8.9 \| 42.1 \|`
GSM8K	math	1.7	[GSM8K]	medium	`\| [GSM8K] \| 8-shot \| 1.7 \| 9.1 \| 44.0 \|`
MATH	math	1.5	[MATH]	medium	`\| [MATH] \| 4-shot \| 1.5 \| 4.5 \| 21.2 \|`

Discussions (0)

none scraped.