HyperAI
HomeNewsLatest PapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAI

AI SOTA Benchmarks

Latest AI model performance metrics, GPU benchmarks, and cutting-edge papers

Categories

Browse tasks by category

Computer Vision

1594 tasks

Image Classification

166 benchmarks, 45 papers

Semantic Segmentation

149 benchmarks, 8 papers

Object Detection

123 benchmarks, 15 papers

Few-Shot Image Classification

90 benchmarks, 47 papers

Image Generation

90 benchmarks, 37 papers

Natural Language Processing

733 tasks

Question Answering

149 benchmarks, 15 papers

Text Classification

85 benchmarks, 13 papers

Machine Translation

83 benchmarks, 15 papers

Named Entity Recognition (NER)

77 benchmarks, 41 papers

Text Generation

71 benchmarks, 14 papers

Medical

288 tasks

Medical Image Segmentation

48 benchmarks, 36 papers

Drug Discovery

28 benchmarks, 28 papers

Sleep Stage Detection

16 benchmarks, 34 papers

Within-Session ERP

15 benchmarks, 1 papers

Text-to-Image Generation

13 benchmarks, 23 papers

Miscellaneous

280 tasks

Image Super-Resolution

68 benchmarks, 10 papers

Recommendation Systems

55 benchmarks, 7 papers

Click-Through Rate Prediction

20 benchmarks, 23 papers

Molecular Property Prediction

18 benchmarks, 29 papers

Age Estimation

16 benchmarks, 42 papers

Methodology

210 tasks

Object Detection

123 benchmarks, 15 papers

Anomaly Detection

76 benchmarks, 29 papers

Classification

71 benchmarks, 49 papers

Domain Adaptation

58 benchmarks, 37 papers

Unsupervised Domain Adaptation

41 benchmarks, 27 papers

Time Series

118 tasks

Time Series Forecasting

86 benchmarks, 49 papers

Time Series Classification

52 benchmarks, 13 papers

Traffic Prediction

33 benchmarks, 11 papers

Pose Estimation

31 benchmarks, 39 papers

GLinear

19 benchmarks, 1 papers

Graphs

112 tasks

Node Classification

127 benchmarks, 42 papers

Link Prediction

80 benchmarks, 36 papers

Graph Classification

72 benchmarks, 7 papers

Image Super-Resolution

68 benchmarks, 10 papers

Graph Regression

17 benchmarks, 40 papers

Audio

87 tasks

Audio Classification

26 benchmarks, 44 papers

Beat Tracking

15 benchmarks, 18 papers

Downbeat Tracking

13 benchmarks, 11 papers

Few-Shot Audio Classification

10 benchmarks, 8 papers

Bandwidth Extension

6 benchmarks, 45 papers

Speech

86 tasks

Speech Recognition

148 benchmarks, 23 papers

Speech Separation

19 benchmarks, 49 papers

Speaker Diarization

15 benchmarks, 12 papers

Speech Emotion Recognition

15 benchmarks, 4 papers

Speech Enhancement

14 benchmarks, 39 papers

Reasoning

75 tasks

Common Sense Reasoning

24 benchmarks, 45 papers

Zero-Shot Video Question Answer

16 benchmarks, 25 papers

Math Word Problem Solving

13 benchmarks, 6 papers

Visual Reasoning

12 benchmarks, 24 papers

3D Human Reconstruction

10 benchmarks, 48 papers

Computer Code

68 tasks

Semantic Segmentation

149 benchmarks, 8 papers

Code Generation

26 benchmarks, 31 papers

Deblurring

16 benchmarks, 4 papers

Text-To-SQL

10 benchmarks, 23 papers

Source Code Summarization

9 benchmarks, 7 papers

Robots

62 tasks

Semantic Segmentation

149 benchmarks, 8 papers

Fraud Detection

12 benchmarks, 15 papers

DeepFake Detection

11 benchmarks, 5 papers

Omniverse Isaac Gym

6 benchmarks, 4 papers

Visual Navigation

6 benchmarks, 2 papers

Knowledge Base

52 tasks

Math Word Problem Solving

13 benchmarks, 6 papers

Text-to-Image Generation

13 benchmarks, 23 papers

Entity Alignment

10 benchmarks, 38 papers

Multi-modal Entity Alignment

8 benchmarks, 18 papers

Document Summarization

7 benchmarks, 46 papers

Playing Games

49 tasks

Continuous Control

73 benchmarks, 21 papers

Image Super-Resolution

68 benchmarks, 10 papers

Atari Games

64 benchmarks, 11 papers

OpenAI Gym

17 benchmarks, 23 papers

SMAC+

16 benchmarks, 18 papers

Music

37 tasks

DeepFake Detection

11 benchmarks, 5 papers

Music Transcription

6 benchmarks, 40 papers

Audio Super-Resolution

4 benchmarks, 22 papers

Cover song identification

4 benchmarks, 18 papers

Music Auto-Tagging

4 benchmarks, 19 papers

Adversarial

34 tasks

Open-Domain Question Answering

15 benchmarks, 30 papers

Face Detection

13 benchmarks, 25 papers

Handwritten Text Recognition

13 benchmarks, 32 papers

Adversarial Defense

10 benchmarks, 34 papers

Adversarial Robustness

7 benchmarks, 5 papers

LLM

15 tasks

multimodal

79 benchmarks, 79 papers

reasoning

57 benchmarks, 61 papers

understanding

48 benchmarks, 47 papers

other

33 benchmarks, 35 papers

knowledge

30 benchmarks, 27 papers

AI Model Performance Benchmarks

Performance metrics of mainstream AI models across various tasks, showcasing the state-of-the-art technology

Image Classification

45 papers | 166 benchmarks

Semantic Segmentation

8 papers | 149 benchmarks

Object Detection

15 papers | 123 benchmarks

Few-Shot Image Classification

47 papers | 90 benchmarks

Image Generation

37 papers | 90 benchmarks

Question Answering

15 papers | 149 benchmarks

Text Classification

13 papers | 85 benchmarks

Machine Translation

15 papers | 83 benchmarks

Named Entity Recognition (NER)

41 papers | 77 benchmarks

Text Generation

14 papers | 71 benchmarks

Medical Image Segmentation

36 papers | 48 benchmarks

Drug Discovery

28 papers | 28 benchmarks

Sleep Stage Detection

34 papers | 16 benchmarks

Within-Session ERP

1 papers | 15 benchmarks

Text-to-Image Generation

23 papers | 13 benchmarks

Image Super-Resolution

10 papers | 68 benchmarks

Recommendation Systems

7 papers | 55 benchmarks

Click-Through Rate Prediction

23 papers | 20 benchmarks

Molecular Property Prediction

29 papers | 18 benchmarks

Age Estimation

42 papers | 16 benchmarks

Object Detection

15 papers | 123 benchmarks

Anomaly Detection

29 papers | 76 benchmarks

Classification

49 papers | 71 benchmarks

Domain Adaptation

37 papers | 58 benchmarks

Unsupervised Domain Adaptation

27 papers | 41 benchmarks

Time Series Forecasting

49 papers | 86 benchmarks

Time Series Classification

13 papers | 52 benchmarks

Traffic Prediction

11 papers | 33 benchmarks

Pose Estimation

39 papers | 31 benchmarks

GLinear

1 papers | 19 benchmarks

Node Classification

42 papers | 127 benchmarks

Link Prediction

36 papers | 80 benchmarks

Graph Classification

7 papers | 72 benchmarks

Image Super-Resolution

10 papers | 68 benchmarks

Graph Regression

40 papers | 17 benchmarks

Audio Classification

44 papers | 26 benchmarks

Beat Tracking

18 papers | 15 benchmarks

Downbeat Tracking

11 papers | 13 benchmarks

Few-Shot Audio Classification

8 papers | 10 benchmarks

Bandwidth Extension

45 papers | 6 benchmarks

Speech Recognition

23 papers | 148 benchmarks

Speech Separation

49 papers | 19 benchmarks

Speaker Diarization

12 papers | 15 benchmarks

Speech Emotion Recognition

4 papers | 15 benchmarks

Speech Enhancement

39 papers | 14 benchmarks

Common Sense Reasoning

45 papers | 24 benchmarks

Zero-Shot Video Question Answer

25 papers | 16 benchmarks

Math Word Problem Solving

6 papers | 13 benchmarks

Visual Reasoning

24 papers | 12 benchmarks

3D Human Reconstruction

48 papers | 10 benchmarks

Semantic Segmentation

8 papers | 149 benchmarks

Code Generation

31 papers | 26 benchmarks

Deblurring

4 papers | 16 benchmarks

Text-To-SQL

23 papers | 10 benchmarks

Source Code Summarization

7 papers | 9 benchmarks

Semantic Segmentation

8 papers | 149 benchmarks

Fraud Detection

15 papers | 12 benchmarks

DeepFake Detection

5 papers | 11 benchmarks

Omniverse Isaac Gym

4 papers | 6 benchmarks

Visual Navigation

2 papers | 6 benchmarks

Math Word Problem Solving

6 papers | 13 benchmarks

Text-to-Image Generation

23 papers | 13 benchmarks

Entity Alignment

38 papers | 10 benchmarks

Multi-modal Entity Alignment

18 papers | 8 benchmarks

Document Summarization

46 papers | 7 benchmarks

Continuous Control

21 papers | 73 benchmarks

Image Super-Resolution

10 papers | 68 benchmarks

Atari Games

11 papers | 64 benchmarks

OpenAI Gym

23 papers | 17 benchmarks

SMAC+

18 papers | 16 benchmarks

DeepFake Detection

5 papers | 11 benchmarks

Music Transcription

40 papers | 6 benchmarks

Audio Super-Resolution

22 papers | 4 benchmarks

Cover song identification

18 papers | 4 benchmarks

Music Auto-Tagging

19 papers | 4 benchmarks

Open-Domain Question Answering

30 papers | 15 benchmarks

Face Detection

25 papers | 13 benchmarks

Handwritten Text Recognition

32 papers | 13 benchmarks

Adversarial Defense

34 papers | 10 benchmarks

Adversarial Robustness

5 papers | 7 benchmarks

multimodal

79 papers | 79 benchmarks

reasoning

61 papers | 57 benchmarks

understanding

47 papers | 48 benchmarks

other

35 papers | 33 benchmarks

knowledge

27 papers | 30 benchmarks

GPU Benchmarks

Latest GPU hardware and software performance evaluations to help you make informed hardware choices

Software Performance

DeepSeek-R1-Distill-Qwen-7B
Environment: vllm
DeepSeek-R1-Distill-Llama-8B
Environment: vllm
DeepSeek-R1-Distill-Qwen-14B
Environment: vllm
DeepSeek-R1-Distill-Qwen-32B
Environment: vllm
DeepSeek-R1-Distill-Llama-70B
Environment: vllm
DeepSeek-R1-Distill-Qwen-7B
Environment: sglang
DeepSeek-R1-Distill-Llama-8B
Environment: sglang
DeepSeek-R1-Distill-Qwen-14B
Environment: sglang
DeepSeek-R1-Distill-Qwen-32B
Environment: sglang
DeepSeek-R1-Distill-Llama-70B
Environment: sglang