Our new X account is live! Follow @wizwand_team for updates

Text-to-Image Retrieval on COCO-CN

78.7R@1

M2-Encoder

Updated 3d ago

Evaluation Results

Method
M2-Encoder 2024.01	78.7	96	98.7	-	-	-
M2-Encoder 2024.01	75.9	95.1	98.3	-	-	-
Wukong_ViT-L 2022.02	74	94.4	98.1	88.6	-	-
M2-Encoder 2024.01	72.5	93.7	97.7	-	-	-
FILIP_ViT-L 2022.02	72.2	92.4	97.2	86.5	-	-
Qwen3-VL-Embedding 2026.02	72.2	-	-	-	-	-
GME 2026.02	71.8	-	-	-	-	-
CLIP_ViT-L 2022.02	70.1	92.2	96.4	86.2	-	-
UME-R1 2026.02	70.1	-	-	-	-	-
FILIP_Swin-L 2022.02	69.9	93.3	97.6	86.7	-	-
ERNIE-ViL 2.0 2022.09	69.6	91.2	96.9	-	86.1	-
ERNIE-ViL 2.0 2024.01	69.6	91.2	96.9	-	-	-
CN-CLIP 2024.01	69.2	89.9	96.1	-	-	-
CLIP_ViT-B 2022.02	68.9	93.3	97.3	86.6	-	-
ObjEmbed 2026.02	68.8	-	-	-	-	-
FG-CLIP2 2026.02	68.1	-	-	-	-	-
Wukong_ViT-B 2022.02	67	91.4	96.7	84.6	-	-
CLIP_Swin-L 2022.02	66.7	91.5	96.3	85.4	-	-
GME 2026.02	66.7	-	-	-	-	-
ObjEmbed 2026.02	66.7	-	-	-	-	-
Wukong_Swin-L 2022.02	66	92.6	97.1	85.5	-	-
MetaCLIP2 2026.02	63.1	-	-	-	-	-
AltCLIP 2024.01	61.3	86	93.2	-	-	-
VLM2Vec-V2 2026.02	60.3	-	-	-	-	-
CN-CLIP 2024.01	57	84.1	93.6	-	-	-
R2D2 2022.09	56.4	85	93.1	-	80.5	-
R2D2 2022.05	56.4	85	93.1	-	-	80.5
R2D2 2024.01	56.4	85	93.1	-	-	-
FILIP_ViT-B 2022.02	56.2	86.8	94.3	76.6	-	-
Taisu 2024.01	53.6	83.7	92.4	-	-	-
Wukong 2022.02	53.4	80.2	90.1	75.1	-	-
Wukong 2022.09	53.4	80.2	90.1	-	75.1	-
Wukong 2022.05	53.4	80.2	90.1	-	-	75.1
Wukong 2024.01	53.4	80.2	90.1	-	-	-
FILIP 2022.02	52.7	79.9	88.6	75.2	-	-
FILIP 2022.05	52.7	79.9	88.6	-	-	75.2
VLM2Vec 2026.02	52.6	-	-	-	-	-
Taiyi 2022.09	51.5	81	90.4	-	-	-
SigLIP2 2026.02	50.7	-	-	-	-	-
FILIP 2022.02	50.5	79.2	88.6	72.1	-	-
Wukong 2022.02	49.2	79.4	87.9	71.9	-	-
CLIP 2022.02	48.7	79.4	88.5	73	-	-
CLIP 2022.02	48.7	76.8	86.4	72.1	-	-
CLIP 2022.05	48.7	76.8	86.4	-	-	72.1
CLIP 2022.02	46.7	78.1	87.7	71.7	-	-
Wukong 2022.02	46.4	77	87.6	70.8	-	-
FILIP 2022.02	37.5	68.1	83	61.8	-	-
BriVL 2022.02	14.8	39	54.2	37.4	-	-
BriVL 2024.01	14.8	39	54.2	-	-	-
M3P 2022.09	-	-	-	-	32.3	-