Qwen2.5-VL-7B-Instruct vs Qwen3-VL-8B-Instruct

Published metadata for 2 models, each read from its own repository.
Field	Qwen2.5-VL-7B-Instruct Qwen/Qwen2.5-VL-7B-Instruct	Qwen3-VL-8B-Instruct Qwen/Qwen3-VL-8B-Instruct
Publisher	Qwen	Qwen
Task	Image and text to text	Image and text to text
Modality	Image and text	Image and text
Parameters, as reported	8.3B parameters	8.8B parameters
Architecture	Qwen2_5_VLForConditionalGeneration	Qwen3VLForConditionalGeneration
Library	transformers	transformers
Context length	128,000 tokens	262,144 tokens
Repository size	16.6 GB	17.5 GB
Artifact formats	safetensors	safetensors
License	apache-2.0	apache-2.0
Access	Open weights, no gate	Open weights, no gate
Memory at 16-bit (weights and margin)	19.9 GB	21 GB
Cheapest GPUs at 16-bit, per hour	1x MI300X, $1.85	1x MI300X, $1.85
Memory at 4-bit (weights and margin)	5 GB	5.3 GB
Cheapest GPUs at 4-bit, per hour	1x MI300X, $1.85	1x MI300X, $1.85
Revision viewed	cc594898137f	0c351dd01ed8
Downloads reported by the hub	6.9M	19.1M
Last observed	2026-09-18	2026-09-18

An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.

Other Reported Results

These results are listed for each model on its own, because the conditions needed to compare them are not stated or do not match. Two results that leave a condition blank are not assumed to share it.

Qwen2.5-VL-7B-Instruct

Benchmark	Conditions	Result	Reported by	Revision	Date
MMMU/MMMU_Pro	Task mmmu_pro_visionMetric mmmu_pro_visionComparison conditions not established	34.3	Model Card Reported by a third party	Evaluated revision not stated	2026-01-28
likaixin/ScreenSpot-Pro	Task android_studio_macosMetric android_studio_macosComparison conditions not established	17.5	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task autocad_windowsMetric autocad_windowsComparison conditions not established	0	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task blender_windowsMetric blender_windowsComparison conditions not established	26.8	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task davinci_macosMetric davinci_macosComparison conditions not established	22.7	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task eviews_windowsMetric eviews_windowsComparison conditions not established	72	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task excel_macosMetric excel_macosComparison conditions not established	20.3	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task fruitloops_windowsMetric fruitloops_windowsComparison conditions not established	7	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task illustrator_windowsMetric illustrator_windowsComparison conditions not established	9.7	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task inventor_windowsMetric inventor_windowsComparison conditions not established	10	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task linux_common_linuxMetric linux_common_linuxComparison conditions not established	20	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task macos_common_macosMetric macos_common_macosComparison conditions not established	36.9	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task matlab_macosMetric matlab_macosComparison conditions not established	33.3	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task origin_windowsMetric origin_windowsComparison conditions not established	9.7	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task overallMetric overallComparison conditions not established	26.8	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task photoshop_windowsMetric photoshop_windowsComparison conditions not established	39.2	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task powerpoint_windowsMetric powerpoint_windowsComparison conditions not established	39	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task premiere_windowsMetric premiere_windowsComparison conditions not established	19.2	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task pycharm_macosMetric pycharm_macosComparison conditions not established	19.2	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task quartus_windowsMetric quartus_windowsComparison conditions not established	15.6	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task solidworks_windowsMetric solidworks_windowsComparison conditions not established	9.1	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task stata_windowsMetric stata_windowsComparison conditions not established	12.2	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task unreal_engine_windowsMetric unreal_engine_windowsComparison conditions not established	45.7	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task vivado_windowsMetric vivado_windowsComparison conditions not established	25	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task vmware_macosMetric vmware_macosComparison conditions not established	58.5	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task vscode_macosMetric vscode_macosComparison conditions not established	32.7	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task windows_common_windowsMetric windows_common_windowsComparison conditions not established	14.8	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17
likaixin/ScreenSpot-Pro	Task word_macosMetric word_macosComparison conditions not established	70.2	ScreenSpot-Pro Leaderboard Reported by a third party	Evaluated revision not stated	2026-03-17

Qwen3-VL-8B-Instruct

Benchmark	Conditions	Result	Reported by	Revision	Date
Delores-Lin/MDPBench	Task arMetric arComparison conditions not established	63.1	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task deMetric deComparison conditions not established	73.7	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task digitalMetric digitalComparison conditions not established	78.4	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task enMetric enComparison conditions not established	71.4	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task esMetric esComparison conditions not established	69.3	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task frMetric frComparison conditions not established	66.2	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task hiMetric hiComparison conditions not established	58.4	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task idMetric idComparison conditions not established	68.5	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task itMetric itComparison conditions not established	79.1	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task jpMetric jpComparison conditions not established	59.9	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task koMetric koComparison conditions not established	61.9	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task latinMetric latinComparison conditions not established	73.6	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task nlMetric nlComparison conditions not established	78.3	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task non_latinMetric non_latinComparison conditions not established	62.5	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task overallMetric overallComparison conditions not established	68.3	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task photographedMetric photographedComparison conditions not established	65	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task ptMetric ptComparison conditions not established	82.2	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task ruMetric ruComparison conditions not established	57.9	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task thMetric thComparison conditions not established	62	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task viMetric viComparison conditions not established	73.4	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task zhMetric zhComparison conditions not established	62.6	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
Delores-Lin/MDPBench	Task zh_tMetric zh_tComparison conditions not established	73.8	MDPBench leaderboard Reported by a third party	Evaluated revision not stated	2026-04-14
llamaindex/ParseBench	Task chartMetric chartSetup Pipeline name: qwen3vl_layoutComparison conditions not established	1.4	ParseBench Reported by a third party	Evaluated revision not stated	2026-04-14
llamaindex/ParseBench	Task layoutMetric layoutSetup Pipeline name: qwen3vl_layoutComparison conditions not established	55.1	ParseBench Reported by a third party	Evaluated revision not stated	2026-04-14
llamaindex/ParseBench	Task meanMetric meanSetup Pipeline name: qwen3vl_layoutComparison conditions not established	46.8	ParseBench Reported by a third party	Evaluated revision not stated	2026-04-14
llamaindex/ParseBench	Task tableMetric tableSetup Pipeline name: qwen3vl_layoutComparison conditions not established	21.4	ParseBench Reported by a third party	Evaluated revision not stated	2026-04-14
llamaindex/ParseBench	Task text_contentMetric text_contentSetup Pipeline name: qwen3vl_layoutComparison conditions not established	89.5	ParseBench Reported by a third party	Evaluated revision not stated	2026-04-14
llamaindex/ParseBench	Task text_formattingMetric text_formattingSetup Pipeline name: qwen3vl_layoutComparison conditions not established	66.8	ParseBench Reported by a third party	Evaluated revision not stated	2026-04-14
tiiuae/PBench	Task averageMetric averageSetup Detection-only model combined with SAM2 to convert boxes to segmentation masks.Comparison conditions not established	49	Community Evals Reported by a third party	Evaluated revision not stated	2026-05-11

SAVRN's Notes on Qwen2.5-VL-7B-Instruct

Where does a model that reads images and text and answers in text belong in a rack? Qwen2.5-VL-7B-Instruct gives three answers: 19.9 GB at 16-bit, 10.0 GB at 8-bit, 5.0 GB at 4-bit, for 8.3B parameters. The cheapest rental listed, a single 192 GB MI300X at $1.85 an hour, sits mostly idle at full precision. So we would co-locate it: run 16-bit beside whatever else the card serves, and hold the rest for a 128,000-token context.

Apache 2.0 lets you modify it, redistribute it and sell what you build; the obligations are keeping notices and stating significant changes. Check the context math: the page lists 128,000 tokens, the configuration shows a 32,768-token sliding window, and the relations point to the YaRN context extension paper, so ask your serving stack which number it honors. Then read the ScreenSpot-Pro rows by application; they span 0.0 to 72.0.

SAVRN Research, 2026-09-18

SAVRN's Notes on Qwen3-VL-8B-Instruct

Put an image and a question in, get text out; at 8.8B parameters that job fits one accelerator. The 16-bit run needs 21.0 GB for 17.5 GB of weights; the cheapest setup is one MI300X with 192 GB at $1.85 an hour on-demand. At 8-bit the need drops to 10.5 GB, at 4-bit to 5.3 GB, so the card decision turns on how much of the 262,144-token context you intend to fill with images rather than on the weights.

Confirm the build first: Qwen ships Dense and MoE variants and a Thinking edition alongside this Instruct one, and the eight MDPBench scores, 58.4 to 78.4, are third-party reported. Apache 2.0 permits commercial use, modification and redistribution, requires keeping the license, copyright notices and NOTICE file and stating significant changes, and carries an express patent grant. No host on the Index prices it, so the price is your own accelerator hour.

SAVRN Research, 2026-09-18

Questions

Which is larger, Qwen2.5-VL-7B-Instruct or Qwen3-VL-8B-Instruct?

Qwen3-VL-8B-Instruct (8.8B parameters) is larger than Qwen2.5-VL-7B-Instruct (8.3B parameters), by the parameter counts their publishers report.

Which is cheaper to run, Qwen2.5-VL-7B-Instruct or Qwen3-VL-8B-Instruct?

At 4-bit, Qwen2.5-VL-7B-Instruct fits on 1x MI300X from $1.85 an hour and Qwen3-VL-8B-Instruct on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use Qwen2.5-VL-7B-Instruct commercially?

Yes. Qwen2.5-VL-7B-Instruct is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Can I use Qwen3-VL-8B-Instruct commercially?

Yes. Qwen3-VL-8B-Instruct is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Search SAVRN