Pixtral-12B-2409

Home/Models/mistralai/Pixtral-12B-2409

mistralai/Pixtral-12B-2409

kind: original
quant: bf16 (16-bit)
family: mistral
modality: —
root model: mistralai/Pixtral-12B-Base-2409
provider: mistralai
license: apache-2.0
language: en, fr, de, es, it, pt, ru, zh, ja
context: —
library: vllm
vRAM (GB): —
size on disk: —
downloads: 6,542
likes: 692
created: —
last modified: 2026-06-02T04:44:58+00:00 (2026)
huggingface: https://huggingface.co/mistralai/Pixtral-12B-2409
tags: vllmmistral-commonenfrdeesitptruzhjabase_model:mistralai/Pixtral-12B-Base-2409base_model:finetune:mistralai/Pixtral-12B-Base-2409license:apache-2.0region:usdeploy:sagemaker

benchmark	category	score	context	conf	evidence
DocVQA	multimodal	90.7	DocVQA (ANLS)	medium	`\| DocVQA (ANLS) \| 90.7 \| <ins>94.5</ins> \| 90.5 \| 84.9 \| 74.4 \|`
MMLU	knowledge	68.5	MMLU (5-shot)	medium	`\| MMLU (5-shot) \| <ins>69.2</ins> \| 68.5 \| 67.9 \| 63.5 \| 63.6 \|`
MathVista	multimodal	54.4	Mathvista (CoT)	medium	`\| Mathvista (CoT) \| <ins>58.0</ins> \| 54.4 \| 36.1 \| 36.4 \| 39.3 \|`
IFEval	instruction-following	50.1	Text IF-Eval	medium	`\| Text IF-Eval \| <ins>61.3</ins> \| 50.1 \| 51.4 \| 50.9 \| 47.4 \|`
MMMU	multimodal	47.6	MMMU (CoT)	medium	`\| MMMU (CoT) \| <ins>52.5</ins> \| 47.6 \| 45.1 \| 40.3 \| 38.3 \|`
IFEval	instruction-following	38.9	MM IF-Eval	medium	`\| MM IF-Eval \| <ins>52.7</ins> \| 38.9 \| 42.5 \| 41.2 \| 31.4 \|`
ChartQA	multimodal	38.6	ChartQA (CoT)	medium	`\| ChartQA (CoT) \| <ins>81.8</ins> \| 38.6 \| 67.1 \| 72.0 \| 67.7 \|`
MATH	math	27.8	Math (Pass@1)	medium	`\| Math (Pass@1) \| <ins>48.1</ins> \| 27.8 \| 38.6 \| 29.2 \| 28.4 \|`
MT-Bench	chat	6.41	Text MT-Bench	medium	`\| Text MT-Bench \| <ins>7.68</ins> \| 6.41 \| 6.94 \| 6.27 \| 6.31 \|`
MT-Bench	chat	5.43	MM MT-Bench	medium	`\| MM MT-Bench \| <ins>6.05</ins> \| 5.43 \| 4.12 \| 3.70 \| 4.46 \|`

none scraped.