List your company
ResearchFormerly WizWand

Image Classification on Country211

86Top-1 AccMLCD
1.614423.522245.4367.3378Aug 4, 2022Mar 9, 2023Oct 13, 2023May 18, 2024Dec 22, 2024Jul 28, 2025Mar 3, 2026
Updated 8mo ago

Evaluation results

MethodLinks
MLCDPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0786
CLIPPre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0785.8
UNICOMPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0785.4
CLIP+Pre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0785.3
OpenCLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0784.3
ITO sub2Pre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0380
PEcore GSize=G2026.0278.2
PEcore G (image only)Size=G, Input=image only2026.0276.7
SigLIP2-g-optSize=g2026.0273.6
DFN-H+2026.0272.5
SLIPPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0370
FLAIRPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0370
ITOPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0370
PEcore LSize=L2026.0267.8
SigLIP2-L/16Backbone=SigLIP2-L, Patch Size=162026.0267
SigLIPPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0360
EVA 18BSize=18B2026.0259.7
LaCLIPArchitecture=ViT-B/16, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0557.9
PEcore BSize=B2026.0257
CLIPArchitecture=ViT-B/16, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0555.1
SigLIP2-B/16Backbone=SigLIP2-B, Patch Size=162026.0254.8
InternVL-C2026.0253.3
SigLIP-L/16Backbone=SigLIP-L, Patch Size=162026.0253.2
LaCLIPArchitecture=ViT-B/32, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0552.2
CLIPPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0350
CLIPArchitecture=ViT-B/32, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0549.7
LOUPEevaluation=linear probing2022.0849.3
LaCLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0546.7
CLIPevaluation=linear probing2022.0846.4
LaSLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0545.4
CLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0544.3
SigLIP-B/16Backbone=SigLIP-B, Patch Size=162026.0244
EVA-CLIP-18BZero-shot=true2024.0243.1
SLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0543
LaCLIPArchitecture=ViT-B/16, Pre-training Data=RedCaps, Protocol=5-way 5-shot2023.0542.6
EVA-CLIP-8BZero-shot=true2024.0241.7
CLIPArchitecture=ViT-B/16, Pre-training Data=RedCaps, Protocol=5-way 5-shot2023.0540.6
MetaCLIP+ViSE2026.0240.137
Only ViSE2026.0240.137
KD 2B to ViT-H, M+V+L4Backbone=ViT-H, Distillation=KD 2B, Target=M+V+L42026.0239.66
DFN5B-CLIP-H/14+Zero-shot=true2024.0237.9
LOUPEmode=zero-shot2022.0837.8
CLIPArchitecture=ViT-B/16, Pre-training Data=CC3M, Protocol=5-way 5-shot2023.0537.3
LaCLIPArchitecture=ViT-B/16, Pre-training Data=CC3M, Protocol=5-way 5-shot2023.0537.3
EVA-02-CLIP-E/14+Zero-shot=true2024.0237
InternVL-CZero-shot=true2024.0235.1
CLIPmode=zero-shot2022.0834.9
OpenCLIP-G/14Zero-shot=true2024.0234.9
DFN5B-CLIP-H/14Zero-shot=true2024.0234.4
KD 2B to ViT-H, ViseBackbone=ViT-H, Distillation=KD 2B, Target=Vise2026.0234.29
CLIP (reported)Pre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0732.7
EVA-01-CLIP-g/14+Zero-shot=true2024.0231.8
CLIPPre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0730.9
CLIPLinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0130.1
DaVinciLinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0129.9
FLAVALinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0128.9
EVA-01-CLIP-g/14Zero-shot=true2024.0228.9
MLCDPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0727.9
xCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.1026.3
X-FM_baseLinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0124.9
CLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.1023.2
FLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0723.1
nCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.1022.7
ITOPre-training Dataset=DataComp-1B, Backbone=ViT-B/16, Training Epochs=10 epochs, Zero-shot protocol=EVA-CLIP2026.0321.8
OpenCLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0721.6
CLIPPre-training Dataset=DataComp-1B, Backbone=ViT-B/16, Training Epochs=10 epochs, Zero-shot protocol=EVA-CLIP2026.0321.4
ITO sub2Pre-training Dataset=DataComp-1B, Backbone=ViT-B/16, Training Epochs=10 epochs, Zero-shot protocol=EVA-CLIP2026.0321
OpenCLIPImage Encoder=ViT-B/16, Zero-shot=true2026.0220.3
ITOBackbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0320.3
CLIPImage Encoder=ViT-B/16, Zero-shot=true2026.0220.2
ITO sub2Backbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0319.1
CLIPBackbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0319
CLIP-MapbaseImage Encoder=ViT-39M/16, Zero-shot=true2026.0218.6
TinyCLIPImage Encoder=ViT-39M/16, Zero-shot=true2026.0218.3
Standard CLIPEvaluation Protocol=zero-shot2024.1115
ITOZero-shot=true, Backbone=ViT-B/16, Pre-training Dataset=Laion100M, Epochs=302026.0313.9
CLIPZero-shot=true, Backbone=ViT-B/16, Pre-training Dataset=Laion100M, Epochs=302026.0313.5
B-cosified RN-50 CLIPPre-training Dataset=CC3M [47], Learning Scheduler=Cyclic, Evaluation Protocol=zero-shot2024.1113
CLIP-MapsmallImage Encoder=ViT-8M/16, Zero-shot=true2026.0212.5
B-cosified RN-50 CLIPPre-training Dataset=ImageNet [16], Learning Scheduler=Cosine, Evaluation Protocol=zero-shot2024.1112
B-cosified RN-50 CLIPPre-training Dataset=ImageNet [16], Learning Scheduler=Cyclic, Evaluation Protocol=zero-shot2024.1112
B-cosified RN-50 CLIPPre-training Dataset=CC3M [47], Learning Scheduler=Cosine, Evaluation Protocol=zero-shot2024.1112
†TinyCLIPImage Encoder=ViT-8M/16, Zero-shot=true2026.0211.7
SLIPZero-shot=true, Backbone=ViT-B/16, Pre-training Dataset=Laion100M, Epochs=302026.0311.5
xCLIPBackbone=ViT-B/16, Pre-trained on=IT35M, Mode=Zero-shot2022.109.3
CLIPBackbone=ViT-B/16, Pre-trained on=IT35M, Mode=Zero-shot2022.109.1
ITO sub2Pre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.037.8
FLAIRPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.036.2
nCLIPBackbone=ViT-B/16, Pre-trained on=IT35M, Mode=Zero-shot2022.105.8
ITOPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.035.7
HyCoCLIPw/ boxes=true2025.105.64
PHyCLIPw/ boxes=true2025.105.56
CLIP-MaptinyImage Encoder=ViT-0.8M/16, Zero-shot=true2026.025.5
SLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.035.4
CLIPw/ boxes=false2025.105.13
CLIPw/ boxes=true2025.105.1
MERUw/ boxes=true2025.105.01
SigLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.035
†TinyCLIPImage Encoder=ViT-0.8M/16, Zero-shot=true2026.024.9
MERUw/ boxes=false2025.104.86