Qwen2.5-VL-32B-Instruct-bnb-4bit

Home/Models/unsloth/Qwen2.5-VL-32B-Instruct-bnb-4bit

unsloth/Qwen2.5-VL-32B-Instruct-bnb-4bit

kind: bnb-4bit
quant: 4bit
family: qwen
modality: image-text-to-text
root model: Qwen/Qwen2.5-VL-32B-Instruct
provider: unsloth
license: apache-2.0
language: en
context: 32,768
library: transformers
vRAM (GB): —
size on disk: —
downloads: 972
likes: 4
created: —
last modified: 2025-05-12T11:02:46+00:00 (2025)
huggingface: https://huggingface.co/unsloth/Qwen2.5-VL-32B-Instruct-bnb-4bit
tags: transformerssafetensorsqwen2_5_vlimage-text-to-textmultimodalunslothconversationalenarxiv:2309.00071arxiv:2502.13923base_model:Qwen/Qwen2.5-VL-32B-Instructbase_model:quantized:Qwen/Qwen2.5-VL-32B-Instructlicense:apache-2.0text-generation-inferenceendpoints_compatible4-bitbitsandbytesdeploy:azureregion:us

benchmark	category	score	context	conf	evidence
DocVQA	multimodal	96.4	DocVQA	medium	`\| DocVQA \| 96.4 \| 96.5 \| 94.8 \|`
MBPP	coding	84.0		high	`\| Qwen2.5-VL-32B \| 78.4 \| 68.8 \| 82.2 \| 46.0 \| 84.0 \| 91.5 \|`
MATH	math	82.2		high	`\| Qwen2.5-VL-32B \| 78.4 \| 68.8 \| 82.2 \| 46.0 \| 84.0 \| 91.5 \|`
MMLU	knowledge	78.4		high	`\| Qwen2.5-VL-32B \| 78.4 \| 68.8 \| 82.2 \| 46.0 \| 84.0 \| 91.5 \|`
MathVista	multimodal	74.8	MathVista	medium	`\| MathVista \| 74.8 \| 70.5 \| 74.7 \|`
MMMU	multimodal	70.2	MMMU	medium	`\| MMMU \| 70.2 \| 64.5 \| 70 \|`
MMLU-Pro	knowledge	68.8		high	`\| Qwen2.5-VL-32B \| 78.4 \| 68.8 \| 82.2 \| 46.0 \| 84.0 \| 91.5 \|`
MMMU	multimodal	51.1	MMMU Pro	medium	`\| MMMU Pro \| 51.1 \| 46.2 \| 49.5 \|`
GPQA Diamond	reasoning	46.0		high	`\| Qwen2.5-VL-32B \| 78.4 \| 68.8 \| 82.2 \| 46.0 \| 84.0 \| 91.5 \|`
OSWorld	agentic	8.83	OSWorld	medium	`\| OSWorld \| 8.83 \| - \| 5.92 \|`
MMBench	multimodal	2.02	MMBench-Video	medium	`\| MMBench-Video \| 2.02 \| 1.7 \| 1.93 \|`

none scraped.