Zero-Shot Video Retrieval on MSR-VTT

Metric: text-to-video R@5 (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	text-to-video R@5▼	Extra Data	Paper	Date↕	Code
1	InternVideo2-6B	78.3	Yes	InternVideo2: Scaling Foundation Models for Mult...	2024-03-22	Code
2	InternVideo2-1B	75.3	Yes	InternVideo2: Scaling Foundation Models for Mult...	2024-03-22	Code
3	VAST, HowToCaption-finetuned	73.2	No	HowToCaption: Prompting LLMs to Transform Video ...	2023-10-07	Code
4	FluxViT-B	71	Yes	Make Your Training Flexible: Towards Deployment-...	2025-03-18	Code
5	LanguageBind(ViT-H/14)	70	Yes	LanguageBind: Extending Video-Language Pretraini...	2023-10-03	Code
6	mPLUG-2	69.7	No	mPLUG-2: A Modularized Multi-modal Foundation Mo...	2023-02-01	Code
7	VAST	68.3	Yes	VAST: A Vision-Audio-Subtitle-Text Omni-Modality...	2023-05-29	Code
8	FluxViT-S	67.5	Yes	Make Your Training Flexible: Towards Deployment-...	2025-03-18	Code
9	LanguageBind(ViT-L/14)	67.5	Yes	LanguageBind: Extending Video-Language Pretraini...	2023-10-03	Code
10	BT-Adapter	64.7	No	BT-Adapter: Video Conversation is Feasible Witho...	2023-09-27	Code
11	UMT-L (ViT-L/16)	64.4	Yes	Unmasked Teacher: Towards Training-Efficient Vid...	2023-03-28	Code
12	vid-TLDR (UMT-L)	63.9	Yes	vid-TLDR: Training Free Token merging for Light-...	2024-03-20	Code
13	Florence	63.8	No	Florence: A New Foundation Model for Computer Vi...	2021-11-22	Code
14	HowToCaption	62	No	HowToCaption: Prompting LLMs to Transform Video ...	2023-10-07	Code
15	ImageBind	61.8	No	ImageBind: One Embedding Space To Bind Them All	2023-05-09	Code
16	HiTeA-17M	60	No	HiTeA: Hierarchical Temporal-Aware Video-Languag...	2022-12-30	-
17	OmniVL	58.4	Yes	OmniVL:One Foundation Model for Image-Language a...	2022-09-15	-
18	CLIP4Clip	57	No	CLIP4Clip: An Empirical Study of CLIP for End to...	2021-04-18	Code
19	Singularity-17M	56.7	Yes	Revealing Single Frame Bias for Video-and-Langua...	2022-06-07	Code
20	Yatai Ji et. al.	54.4	No	Seeing What You Miss: Vision-Language Pre-traini...	2022-11-24	Code
21	HiTeA-5M	54.2	No	HiTeA: Hierarchical Temporal-Aware Video-Languag...	2022-12-30	-
22	Singularity-5M	50.2	Yes	Revealing Single Frame Bias for Video-and-Langua...	2022-06-07	Code
23	Clover	49.5	No	Clover: Towards A Unified Video-Language Alignme...	2022-07-16	Code
24	VIOLET	49.5	No	VIOLET : End-to-End Video-Language Transformers ...	2021-11-24	Code
25	OA-Trans	47.5	No	Object-aware Video-language Pre-training for Ret...	2021-12-01	Code
26	MILES	47.2	No	MILES: Visual BERT Pre-training with Injected La...	2022-04-26	Code
27	FROZEN	46.9	No	Frozen in Time: A Joint Video and Image Encoder ...	2021-04-01	Code
28	Y. Ge et. al.	46.4	No	Bridging Video-text Retrieval with Multiple Choi...	2022-01-13	Code
29	ALPRO	44.7	No	Align and Prompt: Video-and-Language Pre-trainin...	2021-12-17	Code
30	LaT	44.1	No	LaT: Latent Translation with Cycle-Consistency f...	2022-07-11	-
31	A. Nagrani et. al.	39.5	Yes	Learning Audio-Video Modalities from Image Capti...	2022-04-01	-
32	HD-VILA	34.4	No	Advancing High-Resolution Video-Language Represe...	2021-11-19	Code
33	TACo	25	No	TACo: Token-aware Cascade Contrastive Learning f...	2021-08-23	-
34	Norton	24.1	No	Multi-granularity Correspondence Learning from L...	2024-01-30	Code
35	MIL-NCE	24	No	End-to-End Learning of Visual Representations fr...	2019-12-13	Code
36	VideoCLIP	22.2	No	VideoCLIP: Contrastive Pre-training for Zero-sho...	2021-09-28	Code
37	SSML	21.3	No	Noise Estimation Using Density Estimation for Se...	2020-03-06	Code
38	MMT	14.4	Yes	Multi-modal Transformer for Video Retrieval	2020-07-21	Code

#1InternVideo2-6BSOTA
78.3
text-to-video R@5· Extra Data· 2024-03-22
InternVideo2: Scaling Foundation Models for Multimodal Video Understanding Code
#2InternVideo2-1B
75.3
text-to-video R@5· Extra Data· 2024-03-22
InternVideo2: Scaling Foundation Models for Multimodal Video Understanding Code
#3VAST, HowToCaption-finetunedSOTA
73.2
text-to-video R@5· 2023-10-07
HowToCaption: Prompting LLMs to Transform Video Annotations at Scale Code
#4FluxViT-B
71
text-to-video R@5· Extra Data· 2025-03-18
Make Your Training Flexible: Towards Deployment-Efficient Video Models Code
#5LanguageBind(ViT-H/14)SOTA
70
text-to-video R@5· Extra Data· 2023-10-03
LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment Code
#6mPLUG-2SOTA
69.7
text-to-video R@5· 2023-02-01
mPLUG-2: A Modularized Multi-modal Foundation Model Across Text, Image and Video Code
#7VAST
68.3
text-to-video R@5· Extra Data· 2023-05-29
VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset Code
#8FluxViT-S
67.5
text-to-video R@5· Extra Data· 2025-03-18
Make Your Training Flexible: Towards Deployment-Efficient Video Models Code
#9LanguageBind(ViT-L/14)
67.5
text-to-video R@5· Extra Data· 2023-10-03
LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment Code
#10BT-Adapter
64.7
text-to-video R@5· 2023-09-27
BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning Code
#11UMT-L (ViT-L/16)
64.4
text-to-video R@5· Extra Data· 2023-03-28
Unmasked Teacher: Towards Training-Efficient Video Foundation Models Code
#12vid-TLDR (UMT-L)
63.9
text-to-video R@5· Extra Data· 2024-03-20
vid-TLDR: Training Free Token merging for Light-weight Video Transformer Code
#13FlorenceSOTA
63.8
text-to-video R@5· 2021-11-22
Florence: A New Foundation Model for Computer Vision Code
#14HowToCaption
62
text-to-video R@5· 2023-10-07
HowToCaption: Prompting LLMs to Transform Video Annotations at Scale Code
#15ImageBind
61.8
text-to-video R@5· 2023-05-09
ImageBind: One Embedding Space To Bind Them All Code
#16HiTeA-17M
60
text-to-video R@5· 2022-12-30
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training
#17OmniVL
58.4
text-to-video R@5· Extra Data· 2022-09-15
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks
#18CLIP4ClipSOTA
57
text-to-video R@5· 2021-04-18
CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval Code
#19Singularity-17M
56.7
text-to-video R@5· Extra Data· 2022-06-07
Revealing Single Frame Bias for Video-and-Language Learning Code
#20Yatai Ji et. al.
54.4
text-to-video R@5· 2022-11-24
Seeing What You Miss: Vision-Language Pre-training with Semantic Completion Learning Code
#21HiTeA-5M
54.2
text-to-video R@5· 2022-12-30
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training
#22Singularity-5M
50.2
text-to-video R@5· Extra Data· 2022-06-07
Revealing Single Frame Bias for Video-and-Language Learning Code
#23Clover
49.5
text-to-video R@5· 2022-07-16
Clover: Towards A Unified Video-Language Alignment and Fusion Model Code
#24VIOLET
49.5
text-to-video R@5· 2021-11-24
VIOLET : End-to-End Video-Language Transformers with Masked Visual-token Modeling Code
#25OA-Trans
47.5
text-to-video R@5· 2021-12-01
Object-aware Video-language Pre-training for Retrieval Code
#26MILES
47.2
text-to-video R@5· 2022-04-26
MILES: Visual BERT Pre-training with Injected Language Semantics for Video-text Retrieval Code
#27FROZENSOTA
46.9
text-to-video R@5· 2021-04-01
Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval Code
#28Y. Ge et. al.
46.4
text-to-video R@5· 2022-01-13
Bridging Video-text Retrieval with Multiple Choice Questions Code
#29ALPRO
44.7
text-to-video R@5· 2021-12-17
Align and Prompt: Video-and-Language Pre-training with Entity Prompts Code
#30LaT
44.1
text-to-video R@5· 2022-07-11
LaT: Latent Translation with Cycle-Consistency for Video-Text Retrieval
#31A. Nagrani et. al.
39.5
text-to-video R@5· Extra Data· 2022-04-01
Learning Audio-Video Modalities from Image Captions
#32HD-VILA
34.4
text-to-video R@5· 2021-11-19
Advancing High-Resolution Video-Language Representation with Large-Scale Video Transcriptions Code
#33TACo
25
text-to-video R@5· 2021-08-23
TACo: Token-aware Cascade Contrastive Learning for Video-Text Alignment
#34Norton
24.1
text-to-video R@5· 2024-01-30
Multi-granularity Correspondence Learning from Long-term Noisy Videos Code
#35MIL-NCESOTA
24
text-to-video R@5· 2019-12-13
End-to-End Learning of Visual Representations from Uncurated Instructional Videos Code
#36VideoCLIP
22.2
text-to-video R@5· 2021-09-28
VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding Code
#37SSML
21.3
text-to-video R@5· 2020-03-06
Noise Estimation Using Density Estimation for Self-Supervised Multimodal Learning Code
#38MMT
14.4
text-to-video R@5· Extra Data· 2020-07-21
Multi-modal Transformer for Video Retrieval Code