Visual Question Answering Models

Tracked visual question answering models monitored by OpenModelStats. Rankings below compare tracked models against each other — see the methodology for scope and limitations.

Top-10 combined DL 30D
673.5K
Tasks directory position
See all tasks
Growth history
Available
Last refresh
Sep 22, 2026

Trending now

Most downloaded

ModelDL 30D7D
blip-vqa-baseSalesforce527.2K0.7%(up)
vilt-b32-finetuned-vqadandelin75.2K0.4%(up)
deplotgoogle16.3K0.7%(up)
Qwen3.5-2B-MedVLOpenMed13.9K12.6%(up)
blip-vqa-capfilt-largeSalesforce13K0.1%(up)
MiniCPM-V-4-GGUFsecond-state9,29216.1%(up)
MiniCPM-V-2openbmb7,8310.3%(up)
MiniCPM-V-2_6-GGUFsecond-state4,00413.9%(up)
VideoScore2TIGER-Lab3,9342.5%(up)
Gemma4-12B-CVRRdmis-lab2,90821.1%(up)

Fastest growing

Downloads gained over the last seven tracked days.

ModelDL 30D7D
blip-vqa-baseSalesforce527.2K+95.5K
vilt-b32-finetuned-vqadandelin75.2K+27.1K
deplotgoogle16.3K+5,573
blip-vqa-capfilt-largeSalesforce13K+4,255
Qwen3.5-2B-MedVLOpenMed13.9K+3,430
MiniCPM-V-2_6-GGUFsecond-state4,004+3,308
MiniCPM-V-4-GGUFsecond-state9,292+2,283
MiniCPM-V-2openbmb7,831+1,811
VideoScore2TIGER-Lab3,934+1,633
RsEvi-8B-i1-GGUFmradermacher2,329+1,286

New & notable

Published within the last 90 days with meaningful traction.

ModelDL 30D7D
Gemma4-12B-CVRRdmis-lab2,908+339
RsEvi-8B-i1-GGUFmradermacher2,329+1,286
InternVL3-9B-CVRRdmis-lab1,476+17
OpenCaption-2B-VL-SFT-v1.0-i1-GGUFmradermacher1,103+364
OpenCaption-4B-VL-SFT-v1.0-i1-GGUFmradermacher1,013+157
OpenCaption-4B-VL-SFT-v1.0-GGUFmradermacher683+200
Qwen3-VL-8B-CVRRdmis-lab676+24
RsEvi-8B-GGUFmradermacher674+248
Qwen2.5-VL-32B-CVRRdmis-lab539—
Gemma4-31B-CVRRdmis-lab316—