ModelMIEB English · Vision Centric QADimensionsMax inputLanguagesPriceReleased
dDNA VL STEER 2BdnotitiaOpen
71.4
2,048 (or 64)33K tokensen, es, fr +32Aug 2026
TVLM2Vec FullTIGER-LabOpen
62.1
3,072131K tokensenOct 2024
TVLM2Vec LoRATIGER-LabOpen
62
3,072131K tokensenOct 2024
54.3
—0K tokens—Apr 2023
53.4unstated
1,02433K tokens—$0.05 / 1M tokensApr 2026
53.2
—0K tokens—Jan 2024
SBlip2 Opt 2.7BSalesforceOpen
53.1
2,560—enMar 2024
LLanguageBind OmniLanguageBindOpen
53
—0K tokens—Oct 2023
52.8
—0K tokens—Jan 2024
SBlip2 Opt 6.7B CocoSalesforceOpen
52.7
4,096—enMar 2024
52.4
—8K tokens—Jun 2024
SBlip Itm Base FlickrSalesforceOpen
52.4
7681K tokens—Aug 2023
52.3
7680K tokens—Apr 2023
52.2
—0K tokens—Jan 2024
rE5 VroyokongOpen
51.9
—8K tokens—Jul 2024
51.2
—0K tokens—Jan 2024
SBlip Itm Large CocoSalesforceOpen
51.1
7681K tokens—Aug 2023
51.1
7680K tokens—Jan 2024
49.1
7681K tokens—Dec 2023
VVoyage Multimodal 3Voyage AI (MongoDB)
48.6
—33K tokens—$0.12 / 1M tokensNov 2024
47.9unstated
7688K tokens—$0.02 / 1M tokensApr 2026
SBlip Itm Base CocoSalesforceOpen
47.3
7681K tokens—Aug 2023
SBlip Itm Large FlickrSalesforceOpen
47
7681K tokens—Aug 2023
46.9
5120K tokens—Feb 2021
46.7
—2K tokensenJun 2024
46.2
1,1520K tokens—Jan 2024
46.1
—0K tokens—Apr 2023
45.9
7680K tokens—Sep 2022
45.8
1,0240K tokens—Sep 2022
kAlign BasekakaobrainOpen
45.7
7680K tokensenFeb 2023
Jina Clip v1Jina AIOpen
45.4
—8K tokensen$0.05 / 1M tokensMay 2024
45.4
1,0240K tokens—Jan 2024
QEVA02 CLIP B 16QuanSunOpen
45.3unstated
—0K tokens—Apr 2023
44.9
1,0240K tokens—Jan 2024
44.2
—0K tokens—Mar 2023
44.1
7680K tokens—Feb 2021
43.4
7681K tokens—Aug 2023
43.2
1,2800K tokens—Jan 2023
43
5120K tokens—Sep 2022
42.7
5120K tokens—Feb 2021