Image-Text Retrieval on COCO

62.63Recall@1

DetailCLIP-ViT-L/14

Updated 21d ago

Evaluation Results

Method
DetailCLIP-ViT-L/14 2022.08	62.63	-
CLIP-ViT-L/14 2022.08	59.4	-
InternVL-C (MLLM) 2022.08	57.12	-
CLIP-ViT-L/14 2022.08	56.74	-
HQ-CLIP 2026.04	52.2	-
GLIP (detector/grounding-based) 2022.08	50.87	-
MDETR (detector/grounding-based) 2022.08	49.65	-
MetaCLIP-400M 2026.04	48.2	-
BLIP2 (CLIP-like) 2022.08	47.85	-
SLIP (CLIP-like) 2022.08	47.49	-
OpenCLIP 2026.04	46.9	-
DynamiCS 2026.04	46.5	-
DynamiCS 2026.04	46.2	-
RegionCLIP (detector-based) 2022.08	46.06	-
DynamiCS 2026.04	45.1	-
RECLIP* 2026.04	44.4	-
CLIPA* 2026.04	44.4	-
DFN* 2026.04	44.3	-
DynamiCS 2026.04	44.2	-
DetailCLIP-ViT-B/16 2022.08	44.19	-
OpenAI-WIT 2026.04	42.8	-
Random 2026.04	42.7	-
DynamiCS 2026.04	42.5	-
CLIP-ViT-B/16 2022.08	41.24	-
CLIP-ViT-B/16 2022.08	40.93	-
DataComp 2026.04	39.8	-
Random 2026.04	39.6	-
DatologyAI-cls-S/32 2024.11	-	30.2
DatologyAI-ret-S/32 2024.11	-	41.5
TinyCLIP-RN30M 2024.11	-	43.3
TinyCLIP-45M/32 2024.11	-	45.4
TinyCLIP-63M/32 2024.11	-	47.7
MobileCLIP-S0 2024.11	-	49.6
ACED-F0 2024.11	-	51
DatologyAI-cls-B/32 2024.11	-	38.5
DatologyAI-ret-B/32 2024.11	-	49.6
CLIP-KD-RN50 2024.11	-	43.5
OpenAI-RN50 2024.11	-	38.7
OpenAI-CLIP-B/32 2024.11	-	40.3
LAION-CLIP-B/32 2024.11	-	47.7
DataComp-CLIP-B/32 2024.11	-	45.4
MetaCLIP-CLIP-B/32 2024.11	-	46.7
TinyCLIP-39M/16 2024.11	-	46.9
MobileCLIP-S1 2024.11	-	53
ACED-F1 2024.11	-	55.6
OpenAI-RN101 2024.11	-	40.3
MobileCLIP-S2 2024.11	-	54.4
ACED-F2 2024.11	-	58.3
SigLIP 2025.09	-	56.8
DreamLIP 2025.09	-	52.7
LSS 2025.09	-	51.8