Visual Question Answering
Benchmark List
All benchmarks related to this task
clevr
Best model: NS-VQA (1K programs)
Metrics
View Details
clevr-humans
Best model: MDETR
Metrics
View Details
coco-visual-question-answering-vqa-real-2
Best model: HDU-USYD-UNCC
Metrics
View Details
coco-visual-question-answering-vqa-real
Best model: MCB 7 att.
Metrics
View Details
docvqa-test
Best model: Human
Metrics
View Details
docvqa-val
Best model: BERT LARGE Baseline
Metrics
View Details
f-vqa
Best model: ZS-F-VQA
Metrics
View Details
figureqa-test-1
Best model: PReFIL
Metrics
View Details
gqa
Best model: PEVL+
Metrics
View Details
gqa-test-dev
Best model: CFR
Metrics
View Details
gqa-test-std
Best model: ProTo
Metrics
View Details
iconqa
Best model: Patch-TRM
Metrics
View Details
msrvtt-qa
Best model: mPLUG-2
Metrics
View Details
msvd-qa
Best model: mPLUG-2
Metrics
View Details
ok-vqa
Best model: PaLI-X (Single-task FT)
Metrics
View Details
qlevr
Best model: MAC
Metrics
View Details
tdiuc
Best model: Accuracy
Metrics
View Details
textvqa-test-standard
Best model: PaLI
Metrics
View Details
vcr-q-a-dev
Best model: VL-BERTLARGE
Metrics
View Details
vcr-q-ar-dev
Best model: VL-BERTLARGE
Metrics
View Details
vcr-q-ar-test
Best model: GPT4RoI
Metrics
View Details
vcr-qa-r-dev
Best model: VL-BERTLARGE
Metrics
View Details
vcr-qa-r-test
Best model: UNITER (Large)
Metrics
View Details
visual-genome-pairs
Best model: CMN
Metrics
View Details
visual7w
Best model: CMN
Metrics
View Details
vizwiz-2018
Best model: LXR955, No Ensemble
Metrics
View Details
vqa-ce
Best model: RandImg
Metrics
View Details
vqa-cp
Best model: CSS
Metrics
View Details
vqa-v1-test-dev
Best model: SAAA (ResNet)
Metrics
View Details
vqa-v1-test-std
Best model: SAAA (ResNet)
Metrics
View Details
vqa-v2-test-dev
Best model: Oscar
Metrics
View Details
vqa-v2-test-std
Best model: BEiT-3
Metrics
View Details
vqa-v2-val
Best model: BLIP-2 ViT-G FlanT5 XXL (zero-shot)
Metrics
View Details
zs-f-vqa
Best model: SAN † - hard mask
Metrics
View Details
infographicvqa
Best model: Gemini Ultra (pixel only)
Metrics
View Details
hallusionbench
Best model: GPT-4V
Metrics
View Details
autohallusion
Best model: GPT-4V
Metrics
View Details
activitynet
Best model: BLIP-2 T5
Metrics
View Details
artquest
Best model: PrefixLM with CLIP and T5
Metrics
View Details
core-mm
Best model: GPT-4V
Metrics
View Details
dvqa-test-familiar
Best model: PReFIL (Oracle OCR)
Metrics
View Details
egoschema
Best model: Lyra-Pro
Metrics
View Details
retvqa
Best model: MI-BART
Metrics
View Details
a-okvqa
Metrics
View Details
coco-visual-question-answering-vqa-abstract
Metrics
View Details
coco-visual-question-answering-vqa-abstract-1
Metrics
View Details
coco-visual-question-answering-vqa-real-1
Metrics
View Details
gqa-test2019
Metrics
View Details
grit
Metrics
View Details
plotqa-d1
Metrics
View Details
plotqa-d2
Metrics
View Details
tgif-qa
Metrics
View Details
vcr-q-a-test
Metrics
View Details
visual-genome-subjects
Metrics
View Details
vizwiz-2018-answerability
Metrics
View Details
vizwiz-2020-answerability
Metrics
View Details
vizwiz-2020-vqa
Metrics
View Details
vqa-x
Metrics
View Details
ai2d
Metrics
View Details
coco-4
Metrics
View Details
core-mm-1
Metrics
View Details
deepform
Metrics
View Details
docvqa
Metrics
View Details
illusionvqa
Metrics
View Details
imagenet
Metrics
View Details
infoseek
Metrics
View Details
mm-vet
Metrics
View Details
mme
Metrics
View Details
mvbench
Metrics
View Details
ovad-benchmark
Metrics
View Details
pmc-vqa
Metrics
View Details
textvqa
Metrics
View Details
video-mme-1
Metrics
View Details
vlm2-bench
Metrics
View Details
websrc
Metrics
View Details
whoops
Metrics
View Details