,Model,Organization,Publication date,Training dataset,Confidence,Model accessibility,Training code accessibility,Training compute estimation method,Environmental Transparency,Year
753,Super-vector coding,"University of Illinois Urbana-Champaign (UIUC),NEC Laboratories,Rutgers University",1/1/10,"PASCAL VOC 2007,PASCAL VOC 2009",Speculative,,,,None,2010
742,YouTube Video Recommendation System,Google,9/26/10,,,,,,None,2010
743,RNN LM,Johns Hopkins University,9/26/10,WSJ,Speculative,,,Operation counting,None,2010
744,Fisher-Boost,Xerox Research Centre Europe (XRCE),9/5/10,,Unknown,,,,None,2010
745,ReLU (NORB),University of Toronto,6/15/10,,,,,,None,2010
746,ReLU (LFW),University of Toronto,6/15/10,,Unknown,,,,None,2010
752,Stacked Denoising Autoencoders,"University of Montreal / Université de Montréal,University of Toronto",1/3/10,,Unknown,,,,None,2010
748,Deconvolutional Network,New York University (NYU),6/13/10,,Unknown,,,,None,2010
749,Word Representations,"University of Montreal / Université de Montréal,University of Illinois Urbana-Champaign (UIUC)",6/1/10,,,,,,None,2010
750,Feedforward NN,University of Montreal / Université de Montréal,5/13/10,MNIST,,,,Operation counting,None,2010
751,6-layer MLP (MNIST),"IDSIA,University of Lugano,SUPSI",3/1/10,MNIST,Likely,,,Operation counting,None,2010
747,Mid-level Features,"INRIA,Ecole Normale Supèrieure,New York University (NYU)",6/13/10,,Unknown,,,,None,2010
730,HOGWILD!,University of Wisconsin Madison,11/11/11,,Unknown,,,,None,2011
731,NLP from scratch,"NEC Laboratories,Princeton University",11/8/11,,,,,,None,2011
732,Domain Adaptation,University of Maryland,11/6/11,Dataset introduced in 'Adapting Visual Category Models to New Domains',,,,,None,2011
733,Adaptive Subgrad,"Technion - Israel Institute of Technology,Google,University of California (UC) Berkeley",10/3/11,Reuters RCV1,Unknown,,,,None,2011
735,Recursive Neural Network,Stanford University,6/28/11,WSJ,Confident,,,,Indirect,2011
734,Recursive sentiment autoencoder,Stanford University,7/1/11,,Unknown,,,,None,2011
737,Cross-Lingual POS Tagger,"Carnegie Mellon University (CMU),Google Research",6/19/11,,Unknown,,,,None,2011
738,RNN-SpeedUp,"Brno University of Technology,Johns Hopkins University",5/22/11,Penn TreeBank,,,,,None,2011
739,Deep Autoencoders,University of Toronto,4/29/11,,Confident,,,Hardware,Indirect,2011
740,Deep rectifier networks,University of Montreal / Université de Montréal,4/13/11,"CIFAR-10,MNIST,NISTP,NORB",Unknown,,,,None,2011
741,Optimized Single-layer Net,"University of Michigan,Stanford University",4/11/11,,Unknown,,,,None,2011
736,Vector Space Model,Stanford University,6/19/11,IMDb,Confident,,,,Indirect,2011
720,LSTM LM,RWTH Aachen University,9/9/12,,Speculative,,,Operation counting,None,2012
715,DistBelief Vision,Google,12/3/12,ImageNet,Likely,,,,None,2012
716,DistBelief Speech,Google,12/3/12,,Speculative,,,Operation counting,None,2012
717,Bayesian automated hyperparameter tuning,"University of Toronto,University of Sherbrooke,Harvard University",12/2/12,,Unknown,,,,None,2012
718,RNN+LDA+KN5+cache,"Microsoft,Brno University of Technology",12/1/12,Penn TreeBank,,Unreleased,Unreleased,,None,2012
719,AlexNet,University of Toronto,9/30/12,ImageNet,Confident,,,"Operation counting,Hardware,Third-party estimation",Indirect,2012
721,LSTM-300units,RWTH Aachen University,9/1/12,,,Unreleased,Unreleased,,None,2012
724,MV-RNN,Stanford University,7/12/12,,,,,,None,2012
723,Unsupervised High-level Feature Learner,Google,7/12/12,,Likely,,,Operation counting,None,2012
725,Dropout (TIMIT),University of Toronto,6/3/12,TIMIT,,Unreleased,Open (non-commercial),,None,2012
726,Dropout (MNIST),University of Toronto,6/3/12,MNIST,,Unreleased,Open (non-commercial),Operation counting,None,2012
727,Dropout (ImageNet),University of Toronto,6/3/12,ImageNet,,Unreleased,Unreleased,Hardware,None,2012
728,Dropout (CIFAR),University of Toronto,6/3/12,CIFAR-10,,Unreleased,Open (non-commercial),Hardware,None,2012
729,MCDNN (MNIST),IDSIA,2/13/12,MNIST,,,,Operation counting,None,2012
722,Context-dependent RNN,"Microsoft Research,Brno University of Technology",7/27/12,,Unknown,,,,None,2012
698,Visualizing CNNs,New York University (NYU),11/12/13,,,,,"Hardware,Third-party estimation",None,2013
697,TensorReasoner,Stanford University,12/1/13,,Unknown,,,,None,2013
696,DeViSE,Google,12/5/13,,Confident,,,,Indirect,2013
695,TransE,"Universite de Technologie de Compiègne – CNRS,Google",12/5/13,,Speculative,,,Hardware,None,2013
693,RNN for 1B words,Google,12/11/13,One Billion Word benchmark,Speculative,,,,None,2013
690,DOT(S)-RNN,"Aalto University,University of Montreal / Université de Montréal",12/20/13,,,Unreleased,Unreleased,,None,2013
691,DQN,DeepMind,12/19/13,,,,,Operation counting,None,2013
689,Image generation,University of Amsterdam,12/20/13,MNIST,,,,Third-party estimation,None,2013
688,OverFeat,New York University (NYU),12/21/13,,Unknown,,,,None,2013
699,R-CNN (T-net),University of California (UC) Berkeley,11/11/13,,,,,,None,2013
692,Network in Network,National University of Singapore,12/16/13,,Unknown,,,,None,2013
700,Word2Vec (small),Google,10/16/13,,,,,,None,2013
694,DBLSTM,University of Toronto,12/8/13,,,,,,None,2013
702,RNTN,Stanford University,10/1/13,,Likely,Unreleased,Unreleased,,None,2013
713,Textual Imager,Stanford University,1/16/13,,Unknown,,,,None,2013
712,Maxout Networks,University of Montreal / Université de Montréal,2/18/13,,Unknown,,,,None,2013
711,PreTrans-3L-250H,University of Toronto,3/22/13,,,,,,None,2013
710,Selective Search,"University of Trento,University of Amsterdam",4/2/13,,Unknown,,,,None,2013
709,Multilingual DNN,Google,5/26/13,,Confident,,,,Indirect,2013
708,ReLU-Speech,"Google,University of Toronto,New York University (NYU)",5/26/13,,Likely,,,Hardware,None,2013
701,Word2Vec (large),Google,10/16/13,,,,,Third-party estimation,None,2013
707,SemVec,Microsoft Research,6/9/13,,Unknown,,,,None,2013
706,Fisher Vector image classifier,"Universidad Nacional de Cordoba,Inteligent Systems Lab Amsterdam,University of Amsterdam,LEAR Team,INRIA,Xerox Research Centre Europe (XRCE)",6/12/13,ImageNet,,,,Hardware,None,2013
705,RNN+weight noise+dynamic eval,University of Toronto,8/4/13,IAM Online Handwriting Database (IAM-OnDB),,Unreleased,Unreleased,,None,2013
704,Mitosis,IDSIA,9/22/13,,,,,Hardware,None,2013
703,RCTM,University of Oxford,10/1/13,,Likely,,,Hardware,None,2013
714,DistBelief NNLM,Google,1/16/13,,Likely,,,Hardware,None,2013
669,Seq2Seq LSTM,Google,9/10/14,WMT14,,,,"Operation counting,Hardware",None,2014
668,SPN-4+KN5,"Singapore University of Technology & Design,DSO National Laboratories",9/14/14,Penn TreeBank,,Unreleased,Open (non-commercial),,None,2014
667,GoogLeNet / InceptionV1,"Google,University of Michigan,University of North Carolina",9/17/14,"ILSVRC 2014 subset of ImageNet,ImageNet",Confident,,,Third-party estimation,Indirect,2014
666,Deeply-supervised nets,Microsoft Research,9/18/14,"MNIST,CIFAR-10,CIFAR-100,SVHN (Street View House Numbers)",,,,,None,2014
665,Spatially-Sparse CNN,University of Warwick,9/23/14,CIFAR-10,Unknown,,,,None,2014
664,LRCN,"UT Austin,University of Massachusetts Lowell,University of California (UC) Berkeley",11/7/14,TaCoS,,,,,None,2014
661,Cascaded LNet-ANet,Chinese University of Hong Kong (CUHK),11/28/14,"ILSVRC 2012 subset of ImageNet,CelebA",Unknown,,,,None,2014
662,Fully Convolutional Networks,University of California (UC) Berkeley,11/14/14,,Unknown,,,,None,2014
660,SNM-skip,Google,12/3/14,One Billion Word benchmark,Speculative,,,Operation counting,None,2014
659,NTM,Google DeepMind,12/10/14,,Unknown,,,,None,2014
658,Fractional Max-Pooling,University of Warwick,12/18/14,CIFAR-100,Likely,,,Hardware,None,2014
670,Large regularized LSTM,"New York University (NYU),Google Brain",9/8/14,Penn TreeBank,,Unreleased,Open source,,None,2014
656,DeepLab,"Google,University of California Los Angeles (UCLA)",12/22/14,,Unknown,,,,None,2014
663,SC-NLM,University of Toronto,11/10/14,"COCO,Flickr30K Entities",Confident,,,,Indirect,2014
657,ADAM (CIFAR-10),"University of Amsterdam,OpenAI,University of Toronto",12/22/14,,,,,Third-party estimation,None,2014
671,VGG19,University of Oxford,9/4/14,ILSVRC 2012 subset of ImageNet,,,,,None,2014
673,RNNsearch-50*,"Jacobs University Bremen,University of Montreal / Université de Montréal",9/1/14,WMT'14 + selection,,,,Third-party estimation,None,2014
672,VGG16,University of Oxford,9/4/14,ILSVRC 2012 subset of ImageNet,Confident,,,Hardware,Indirect,2014
686,GloVe (32B),Stanford University,1/1/14,Common Crawl,,,,,None,2014
685,HyperNEAT,University of Texas at Austin,3/5/14,,,,,,None,2014
684,Paragraph Vector,Google,5/14/14,IMDb,Confident,,,,Indirect,2014
683,AdaRNN,Beihang University,6/1/14,,Confident,,,,Indirect,2014
682,GRUs,"University of Montreal / Université de Montréal,Jacobs University,University of Maine",6/3/14,,Unknown,,,,None,2014
681,Two-stream ConvNets for action recognition,University of Oxford,6/9/14,,Unknown,,,,None,2014
687,GloVe (6B),Stanford University,1/1/14,Gigaword5 + Wikipedia2014,,,,,None,2014
679,SPPNet,"Microsoft,Xi’an Jiaotong University,University of Science and Technology of China",6/18/14,ImageNet-1k,,,,Hardware,None,2014
678,Fragment embedding,Stanford University,6/21/14,Flickr30K Entities,Likely,,,,None,2014
677,RNN-WER,"DeepMind,University of Toronto",6/22/14,WSJ,Likely,,,,None,2014
676,DeepFace,"Tel Aviv University,Facebook",6/23/14,,Unknown,,,,None,2014
675,Multiresolution CNN,"Google,Stanford University",6/23/14,,,,,,None,2014
674,SmooCT,University College London (UCL),7/1/14,,,,,Hardware,None,2014
680,GANs,University of Montreal / Université de Montréal,6/10/14,CIFAR-10,Speculative,,,Third-party estimation,None,2014
638,AlphaGo Fan,DeepMind,10/1/15,,,Unreleased,Unreleased,Hardware,None,2015
637,Multi-scale Dilated CNN,"Princeton University,Intel Labs",11/23/15,,Unknown,,,,None,2015
636,Netflix Recommender System,Netflix,12/1/15,,Unknown,,,,None,2015
635,Inception v3,"Google,University College London (UCL)",12/2/15,ILSVRC 2012 subset of ImageNet,,,,,None,2015
634,DeepSpeech2 (English),Baidu Research - Silicon Valley AI Lab,12/8/15,,Confident,,,"Operation counting,Third-party estimation",Indirect,2015
630,BPL,"University of Toronto,New York University (NYU),Massachusetts Institute of Technology (MIT)",12/11/15,,Unknown,,,,None,2015
632,ResNet-110 (CIFAR-10),Microsoft,12/10/15,,,,,,None,2015
631,ResNet-152 (ImageNet),Microsoft,12/10/15,ILSVRC 2012 subset of ImageNet,,,,Operation counting,None,2015
629,Advantage Learning,Google DeepMind,12/15/15,,Unknown,,,,None,2015
628,"Variational (untied weights, MC) LSTM (Large)",University of Cambridge,12/16/15,,,Unreleased,Unreleased,,None,2015
639,Deep Deterministic Policy Gradients,Google DeepMind,9/9/15,,Unknown,,,,None,2015
633,SSD,,12/8/15,,Confident,Open weights (unrestricted),,,Indirect,2015
640,BPE,University of Edinburgh,8/31/15,WMT'15,,,,,None,2015
647,Trajectory-pooled conv nets,"Chinese University of Hong Kong (CUHK),Chinese Academy of Sciences",5/19/15,"ImageNet,UCF101",,,,,None,2015
642,"Listen, Attend and Spell","Google,Carnegie Mellon University (CMU)",8/20/15,,Unknown,Unreleased,Unreleased,,None,2015
641,LSTM-Char-Large,"Harvard University,New York University (NYU)",8/26/15,Penn TreeBank,,Unreleased,Open source,,None,2015
654,CRF-RNN,"University of Oxford,Stanford University,Baidu",2/11/15,,Unknown,,,,None,2015
652,DQN-2015,Google,2/25/15,,,,,,None,2015
651,Constituency-Tree LSTM,"MetaMind Inc,Stanford University",2/28/15,,,,,,None,2015
650,genCNN + dyn eval,"Chinese Academy of Sciences,Huawei Noah's Ark Lab,Dublin City University",3/17/15,Penn TreeBank,,Unreleased,Unreleased,,None,2015
649,Fast R-CNN,Microsoft Research,4/30/15,,Unknown,,,,None,2015
653,TRPO,University of California (UC) Berkeley,2/19/15,,Confident,Unreleased,,,Indirect,2015
655,"MSRA (C, PReLU)",Microsoft Research,2/6/15,,,,,Hardware,None,2015
646,Faster R-CNN,Microsoft Research,6/4/15,,Unknown,Open weights (unrestricted),Open source,,Indirect,2015
645,YOLO,"University of Washington,Allen Institute for AI,Facebook AI Research",6/8/15,,,,,,None,2015
644,BatchNorm,Google,6/15/15,ImageNet,Confident,,,,Indirect,2015
643,Search-Proven Best LSTM,Google,7/6/15,,,Unreleased,Unreleased,,None,2015
648,Deep LSTM video classifier,"University of Texas at Austin,Google",5/1/15,,Unknown,,,,None,2015
591,BIDAF,"University of Washington,Allen Institute for AI",11/5/16,"SQuAD,DMQA,GloVe",Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2016
600,TSN,"ETH Zurich,Shenzhen Institute of Advanced Technology,Chinese University of Hong Kong (CUHK)",9/17/16,,Unknown,,,,None,2016
599,Wide Residual Network,Université Paris-Est,9/19/16,,Unknown,,,,None,2016
598,GNMT,Google,9/26/16,,,Hosted access (no API),Unreleased,"Hardware,Third-party estimation",None,2016
597,Pointer Sentinel-LSTM (medium),"MetaMind Inc,Salesforce",9/26/16,Penn TreeBank,,Unreleased,Unreleased,,None,2016
596,Zoneout + Variational LSTM (WT2),"MetaMind Inc,Salesforce",9/26/16,WikiText-2,,Unreleased,Unreleased,,None,2016
594,Differentiable neural computer,Google DeepMind,10/12/16,,Unknown,,,,None,2016
593,SPIDER2,"Griffith University,University of Iowa,Dezhou University",10/28/16,Unspecified,Likely,Open weights (non-commercial),,Operation counting,Indirect,2016
592,VD-LSTM+REAL Large,"Salesforce Research,Stanford University",11/4/16,Penn TreeBank,,Unreleased,Unreleased,,None,2016
590,NAS with base 8 and shared embeddings,Google Brain,11/5/16,Penn TreeBank,,Unreleased,Unreleased,,None,2016
583,Elastic weight consolidation,DeepMind,12/2/16,,Unknown,,,,None,2016
588,Deeply-recursive ConvNet,Seoul National University,11/11/16,,Unknown,,,,None,2016
587,ResNeXt-50,"University of California San Diego,Facebook",11/16/16,,,,,,None,2016
586,PolyNet,Chinese University of Hong Kong (CUHK),11/17/16,ImageNet,Likely,,,"Comparison with other models,Operation counting",None,2016
585,RefineNet,"University of Adelaide,Australian Centre for Robotic Vision",11/20/16,,Unknown,,,,None,2016
584,Image-to-image cGAN,University of California (UC) Berkeley,11/21/16,,Unknown,,,,None,2016
601,Stacked hourglass network,University of Michigan,9/17/16,,Unknown,,,,None,2016
582,PointNet,Stanford University,12/2/16,,Unknown,,,,None,2016
581,GAN-Advancer,OpenAI,12/5/16,,Unknown,Unreleased,Open (non-commercial),,None,2016
580,Diabetic Retinopathy Detection Net,"UT Austin,University of California (UC) Berkeley,Google",12/13/16,,Unknown,,,,None,2016
579,GCNN-14,Facebook AI Research,12/23/16,WikiText-103,Unknown,Unreleased,Unreleased,,None,2016
578,YOLOv2,"University of Washington,Allen Institute for AI",12/25/16,,,Open weights (non-commercial),Unreleased,,Indirect,2016
589,NASv3 (CIFAR-10),Google Brain,11/5/16,,Likely,,,"Third-party estimation,Operation counting",None,2016
602,ResNet-1001,Microsoft,9/17/16,"CIFAR-10,CIFAR-100",,,,,None,2016
595,Xception,Google,10/7/16,JFT,Confident,,,Hardware,Indirect,2016
604,MS-CNN,"IBM,University of California San Diego",9/17/16,,Unknown,,,,None,2016
603,ResNet-200,Microsoft Research Asia,9/17/16,ImageNet,Speculative,Unreleased,Open (non-commercial),Hardware,None,2016
627,AlphaGo Lee,DeepMind,1/27/16,,Speculative,Unreleased,Unreleased,Comparison with other models,None,2016
626,Convolutional Pose Machines,Carnegie Mellon University (CMU),1/30/16,,Unknown,,,,None,2016
625,A3C FF hs,"Google,University of Montreal / Université de Montréal",2/4/16,,Unknown,,,,None,2016
624,Inception-ResNet-V2,Google,2/23/16,,,,,,None,2016
623,Inceptionv4,Google,2/23/16,,,,,,None,2016
621,Binarized Neural Network (MNIST),"Technion - Israel Institute of Technology,Columbia University,University of Montreal / Université de Montréal",3/17/16,MNIST,Speculative,,,,None,2016
620,Symmetric Residual Encoder-Decoder Net,"Nanjing University,University of Adelaide",3/30/16,,Unknown,,,,None,2016
619,Gated HORNN (3rd order),York University,4/30/16,Penn TreeBank,,Unreleased,Unreleased,,None,2016
618,Named Entity Recognition model,Carnegie Mellon University (CMU),5/29/16,CoNLL2003,Confident,,,Hardware,Indirect,2016
617,Part-of-sentence tagging model,Carnegie Mellon University (CMU),5/29/16,"WSJ,Penn TreeBank",Confident,,,Hardware,Indirect,2016
622,SqueezeNet,"DeepScale,University of California (UC) Berkeley,Stanford University",2/24/16,,,,,,None,2016
615,DMN,Salesforce,6/20/16,,Unknown,,,,None,2016
605,Youtube recommendation model,Google,9/15/16,,Unknown,,,,None,2016
616,Spatiotemporal fusion ConvNet,"Graz University of Technology,University of Oxford",6/1/16,UCF101,,,,,None,2016
606,WaveNet,Google DeepMind,9/12/16,,Unknown,,,,None,2016
607,Multi-task Cascaded CNN,"Chinese Academy of Sciences,Chinese University of Hong Kong (CUHK)",8/26/16,,Unknown,,,,None,2016
609,SimpleNet,"Sensifai,Islamic Azad University,Technicolor R&I,Institute for Research in Fundamental Sciences (IPM)",8/22/16,"CIFAR-10,ImageNet",Confident,,,,Indirect,2016
608,DenseNet-264,"Tsinghua University,Facebook AI Research,Cornell University",8/25/16,,,,,,None,2016
611,VD-RHN,"ETH Zurich,IDSIA",7/12/16,Penn TreeBank,,Unreleased,Open source,,None,2016
612,fastText,Facebook AI Research,7/6/16,,Unknown,,,,None,2016
613,Wide & Deep,Google,6/24/16,,Unknown,,,,None,2016
614,R-FCN,"Tsinghua University,Microsoft Research",6/21/16,"PASCAL VOC 2007,PASCAL VOC 2012,COCO",,,,Hardware,None,2016
610,Character-enriched word2vec,Facebook AI Research,7/15/16,,Unknown,,,,None,2016
546,Cutout-regularized net,"University of Guelph,Vector Institute,CIFAR AI Research",8/15/2017,,Unknown,,,,None,2017
538,LSTM + dynamic eval,University of Edinburgh,9/21/2017,WikiText-2,,Unreleased,Open source,,None,2017
536,AlphaGo Zero,DeepMind,10/18/2017,,,Unreleased,Unreleased,"Third-party estimation,Hardware",None,2017
537,AWD-LSTM+WT+Cache+IOG (WT2),NTT Communication Science Laboratories,9/26/2017,,,Unreleased,Open (non-commercial),,None,2017
539,ISS,"Duke University,Microsoft",9/15/2017,,,Unreleased,Open source,,None,2017
544,Adversarial Joint Adaptation Network (ResNet),"Tsinghua University,University of California (UC) Berkeley",8/17/2017,"Office-31,ILSVRC 2012 subset of ImageNet",Speculative,,,,None,2017
541,SENet (ImageNet),"Chinese Academy of Sciences,University of Oxford",9/5/2017,ImageNet,,,,,None,2017
542,GL-LWGC-AWD-MoS-LSTM + dynamic evaluation (WT2),Ben-Gurion University of the Negev,8/29/2017,WikiText-2,,Unreleased,Unreleased,,None,2017
543,Libratus,Carnegie Mellon University (CMU),8/19/2017,,,Unreleased,Unreleased,Hardware,None,2017
545,NeuMF (Pinterest),"Shandong University,Texas A&M,National University of Singapore,Columbia University",8/16/2017,,Unknown,,,,None,2017
535,AlphaGo Master,DeepMind,10/19/2017,,,Unreleased,Unreleased,Benchmarks,None,2017
540,PyramidNet,Korea Advanced Institute of Science and Technology (KAIST),9/6/2017,"CIFAR-10,CIFAR-100",Likely,Open weights (unrestricted),Open source,Operation counting,Indirect,2017
534,LRSO-GAN,University of Technology Sydney,10/22/2017,,Unknown,,,,None,2017
522,2-layer-LSTM+Deep-Gradient-Compression,"Tsinghua University,Stanford University,NVIDIA",12/5/2017,,,Unreleased,Unreleased,,None,2017
532,CapsNet (MultiMNIST),Google Brain,10/26/2017,,,,,,None,2017
531,ProgressiveGAN,NVIDIA,10/27/2017,,Unknown,,,,None,2017
530,PhraseCond,"Carnegie Mellon University (CMU),University of Pittsburgh",10/28/2017,SQuAD 1.1,Confident,,,,Indirect,2017
529,S-Norm,"University of Washington,Allen Institute for AI",10/29/2017,TriviaQA,Confident,,,,Indirect,2017
528,DCN+,Salesforce Research,10/31/2017,SQuAD,Confident,Unreleased,,,Indirect,2017
527,Fraternal dropout + AWD-LSTM 3-layer (WT2),"Jagiellonian University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),University of Montreal / Université de Montréal",10/31/2017,WikiText-2,,Unreleased,Open source,,None,2017
526,"AWD-LSTM-MoS + dynamic evaluation (WT2, 2017)",Carnegie Mellon University (CMU),11/10/2017,,,Unreleased,Open source,,None,2017
525,TriNet,"Visual Computing Institute,RWTH Aachen University",11/21/2017,,Unknown,,,,None,2017
524,PNAS-net,"Johns Hopkins University,Google AI,Stanford University",12/2/2017,,,,,,None,2017
523,PNASNet-5,"Johns Hopkins University,Google AI,Stanford University",12/2/2017,ImageNet-1k,,,,Comparison with other models,None,2017
521,AlphaZero,DeepMind,12/5/2017,,,Unreleased,Unreleased,Third-party estimation,None,2017
520,Tacotron 2,"Google,University of California (UC) Berkeley",12/19/2017,,Confident,,,,Indirect,2017
533,CapsNet (MNIST),Google Brain,10/26/2017,MNIST,,,,,None,2017
547,EI-REHN-1000D,Korea Advanced Institute of Science and Technology (KAIST),8/14/2017,,,Unreleased,Unreleased,,None,2017
561,HRA,"Maluuba,Microsoft",6/13/2017,,Unknown,,,,None,2017
549,RetinaNet-R101,Facebook AI Research,8/7/2017,COCO,,,,Hardware,None,2017
548,OpenAI TI7 DOTA 1v1,OpenAI,8/11/2017,,,,,Third-party estimation,None,2017
577,DeepStack,"University of Alberta,Charles University,Czech Technical University",1/6/2017,,Speculative,,,Hardware,None,2017
576,OR-WideResNet,"Duke University,University of Chinese Academy of Sciences",1/7/2017,CIFAR-10,Confident,,,,Indirect,2017
575,MoE-Multi,"Jagiellonian University,Google Brain",1/23/2017,,,Unreleased,,Hardware,None,2017
574,DnCNN,"Harbin Institute of Technology,Hong Kong Polytechnic University,ULSee Inc.,Xi’an Jiaotong University",2/1/2017,,Unknown,,,,None,2017
573,Prototypical networks,"University of Toronto,Twitter",3/15/2017,,Unknown,,,,None,2017
572,Mask R-CNN,Facebook AI Research,3/30/2017,COCO,Unknown,,,,None,2017
571,WGAN-GP,"Courant Institute of Mathematical Sciences,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms)",3/31/2017,,Unknown,,,,None,2017
570,MobileNet,Google,4/17/2017,,,,,,None,2017
569,DeepLab (2017),"Johns Hopkins University,Google,University College London (UCL)",4/27/2017,,Unknown,,,,None,2017
568,Mnemonic Reader,"Fudan University,Microsoft Research",5/8/2017,SQuAD,Confident,,,,Indirect,2017
567,SRGAN,Twitter,5/25/2017,,Unknown,Unreleased,Unreleased,,None,2017
566,Inflated 3D ConvNet,"DeepMind,University of Oxford",6/1/2017,,Unknown,,,,None,2017
565,PointNet++,Stanford University,6/7/2017,,Unknown,,,,None,2017
564,Reading Twice for NLU,DeepMind,6/8/2017,"TriviaQA,SQuAD",Unknown,,,,None,2017
563,EDSR,Seoul National University,6/10/2017,,Unknown,,,,None,2017
550,RetinaNet-R50,Facebook AI Research,8/7/2017,,,,,,None,2017
552,GSM,"Peking University,Microsoft Research",7/30/2017,SQuAD,Likely,,,,None,2017
553,ConvS2S (ensemble of 8 models),Meta AI,7/25/2017,"WMT English-German,WMT14,Gigaword",Likely,,,Hardware,None,2017
554,PSPNet,Chinese University of Hong Kong (CUHK),7/21/2017,,Unknown,,,,None,2017
555,NASNet-A,Google Brain,7/21/2017,,,,,,None,2017
551,AWD-LSTM - 3-layer LSTM (tied) + continuous cache pointer (WT2),Salesforce Research,8/7/2017,WikiText-2,,Unreleased,Open source,,None,2017
557,JFT,"Google Research,Carnegie Mellon University (CMU)",7/10/2017,JFT-300M,Confident,,,Hardware,Indirect,2017
558,ShuffleNet v1,Megvii Inc,7/3/2017,,,,,,None,2017
559,NoisyNet-Dueling,DeepMind,6/30/2017,,Unknown,Unreleased,Unreleased,,None,2017
560,DeepLabV3,Google,6/17/2017,,Unknown,,,,None,2017
562,Transformer,"Google Research,Google Brain",6/12/2017,"WMT English-German,WMT14",Confident,Unreleased,Unreleased,Hardware,Indirect,2017
556,AWD-LSTM,"DeepMind,University of Oxford",7/18/2017,WikiText-2,,Unreleased,Unreleased,,None,2017
483,Transformer + Simple Recurrent Unit,"ASAPP,Cornell University,Google,Princeton University",9/17/2018,WMT English-German,Confident,Unreleased,Unreleased,Hardware,Indirect,2018
484,ESRGAN,"Chinese University of Hong Kong (CUHK),Chinese Academy of Sciences,Nanyang Technological University",9/1/2018,"DIV2K,Flickr2K,OutdoorSceneTraining (OST)",Unknown,,,,None,2018
485,(ensemble): AWD-LSTM-DOC (fin) √ó 5 (WT2),"NTT Communication Science Laboratories,Tohoku University",8/30/2018,WikiText-2,,Open weights (unrestricted),Open source,,Indirect,2018
486,Big Transformer for Back-Translation,"Facebook AI Research,Google Brain",8/28/2018,WMT English-German,Likely,Open weights (unrestricted),Open source,Hardware,Indirect,2018
489,Big-Little Net,IBM,7/10/2018,ImageNet,Likely,Open weights (unrestricted),Open source,Operation counting,Indirect,2018
488,Big-Little Net (speech),IBM,7/10/2018,"Switchboard,Fisher",Speculative,Open weights (unrestricted),Open source,Operation counting,Indirect,2018
490,RCAN,Northeastern University,7/8/2018,DIV2K,Unknown,,,,None,2018
491,Population-based DRL,DeepMind,7/3/2018,,,Unreleased,Unreleased,Third-party estimation,None,2018
481,LSTM+NeuralCache,"KU Leuven,ESAT - PSI,Apple",9/24/2018,,,Unreleased,Unreleased,,None,2018
487,AWD-LSTM-MoS+PDR + dynamic evaluation (WT2),IBM,8/14/2018,WikiText-2,,Unreleased,Unreleased,,None,2018
480,BigGAN-deep 512x512,"Heriot-Watt University,DeepMind",9/28/2018,JFT-300M,Likely,Open weights (unrestricted),Unreleased,Third-party estimation,Indirect,2018
474,Mesh-TensorFlow Transformer 2.9B (translation),Google Brain,11/5/2018,WMT14,Likely,Unreleased,Open source,Hardware,None,2018
478,BERT-Large,Google,10/11/2018,,,Open weights (unrestricted),Open source,"Operation counting,Hardware",Indirect,2018
477,MetaMimic,Google,10/11/2018,,,,,,None,2018
476,TrellisNet,"Carnegie Mellon University (CMU),Bosch Center for Artificial Intelligence,Intel Labs",10/15/2018,WikiText-103,,Unreleased,Open source,,None,2018
475,MemoReader,"Samsung,Korea University",10/31/2018,TriviaQA,Unknown,Unreleased,,,None,2018
492,ShuffleNet v2,"Tsinghua University,Megvii Inc",6/30/2018,,,,,,None,2018
473,Mesh-TensorFlow Transformer 4.9B (language),Google Brain,11/5/2018,"Wikipedia,One Billion Word benchmark",Confident,Unreleased,Open source,Hardware,Indirect,2018
472,Fine-tuned-AWD-LSTM-DOC (fin),Samsung R&D Institute Russia,11/12/2018,Penn TreeBank,Confident,Unreleased,Unreleased,Operation counting,Indirect,2018
471,Multi-cell LSTM,University of Hyderabad,11/15/2018,,,Unreleased,Unreleased,,None,2018
470,GPipe (Amoeba),Google,11/16/2018,ImageNet,,,,,None,2018
469,GPipe (Transformer),Google,11/16/2018,,,,,,None,2018
479,Transformer (Adaptive Input Embeddings) WT103,Facebook AI Research,9/28/2018,WikiText-103,Confident,Open weights (unrestricted),Open source,"Hardware,Operation counting",Indirect,2018
493,QT-Opt,"Google Brain,University of California (UC) Berkeley",6/27/2018,,Likely,Unreleased,,Hardware,None,2018
482,"AWD-LSTM-MoS + dynamic evaluation (WT2, 2018)","Peking University,Microsoft Research Asia",9/18/2018,WikiText-2,,Unreleased,Open (non-commercial),,None,2018
495,MobileNetV2,Google,6/18/2018,,,,,,None,2018
519,Refined Part Pooling,"Tsinghua University,University of Technology Sydney,University of Texas at San Antonio",1/9/2018,"ImageNet-1k,Market-1501",Confident,,,Hardware,Indirect,2018
494,DARTS,"DeepMind,Carnegie Mellon University (CMU)",6/24/2018,WikiText-2,,Unreleased,Open source,,None,2018
518,ULM-FiT,"University of San Francisco,Insight Centre NUI Galway,Fast.ai",1/18/2018,"IMDb,Yelp,Trec-6,DBpedia,AG news,WikiText-103",Speculative,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2018
517,ELMo,"University of Washington,Allen Institute for AI",2/1/2018,,Speculative,,,Third-party estimation,None,2018
516,QRNN,Salesforce Research,2/1/2018,WikiText-103,,Unreleased,Unreleased,,None,2018
515,AmoebaNet-A (F=190),Google Brain,2/5/2018,,,,,,None,2018
513,IMPALA,DeepMind,2/5/2018,,,Unreleased,Open source,Third-party estimation,None,2018
512,DeepLabV3+,Google,2/7/2018,"ImageNet-1k,COCO,JFT-300M",Unknown,,,,None,2018
511,ENAS,"Google Brain,Carnegie Mellon University (CMU),Stanford University",2/9/2018,Penn TreeBank,,Unreleased,Open source,,None,2018
510,TCN (P-MNIST),"Carnegie Mellon University (CMU),Intel Labs",2/15/2018,P-MNIST,Confident,,,,Indirect,2018
509,Spectrally Normalized GAN,"Preferred Networks Inc,Ritsumeikan University,National Institute of Informatics",2/16/2018,CIFAR-10,Unknown,,,,None,2018
508,Residual Dense Network,"Northeastern University,University of Rochester",2/24/2018,DIV2K,Unknown,,,,None,2018
514,AmoebaNet-A (F=448),Google Brain,2/5/2018,ImageNet-1k,,Unreleased,Unreleased,Hardware,None,2018
506,LSTM (2018),"Intel Labs,Carnegie Mellon University (CMU)",3/4/2018,Penn TreeBank,,Open weights (unrestricted),Open source,,Indirect,2018
497,GPT-1,OpenAI,6/1/2018,"BookCorpus (BooksCorpus, Toronto Book Corpus)",,Open weights (unrestricted),Open source,Operation counting,Indirect,2018
507,Chinese - English translation,Microsoft,3/1/2018,,Unknown,,,,None,2018
498,aLSTM(depth-2)+RecurrentPolicy (WT2),"University of Manchester,Alan Turing Institute",5/22/2018,,,Unreleased,Open source,,None,2018
496,Relational Memory Core,"DeepMind,University College London (UCL)",6/5/2018,WikiText-103,Unknown,Unreleased,Unreleased,,None,2018
500,ResNeXt-101 32x48d,Facebook,5/2/2018,"ImageNet,Instagram",Confident,Open weights (non-commercial),Unreleased,Operation counting,Indirect,2018
501,Diffractive Deep Neural Network,University of California Los Angeles (UCLA),4/14/2018,MNIST,Likely,,,,None,2018
499,Dropout-LSTM+Noise(Bernoulli) (WT2),"Columbia University,New York University (NYU),Princeton University",5/3/2018,,,Unreleased,Unreleased,,None,2018
502,YOLOv3,University of Washington,4/8/2018,ImageNet,,Unreleased,Unreleased,Operation counting,None,2018
503,"LSTM (Hebbian, Cache, MbPA)","DeepMind,University College London (UCL)",3/27/2018,Project Gutenberg,Confident,Unreleased,Unreleased,"Hardware,Operation counting",Indirect,2018
504,4 layer QRNN (h=2500),Salesforce Research,3/22/2018,WikiText-103,,Unreleased,Open source,,None,2018
505,Rotation,École des Ponts ParisTech,3/21/2018,CIFAR-10,,,,,None,2018
418,DistilBERT,Hugging Face,10/2/2019,"Wikipedia,BookCorpus (BooksCorpus, Toronto Book Corpus)",,Open weights (unrestricted),Open source,Hardware,Indirect,2019
419,AlphaX-1,"Facebook AI Research,Brown University",10/2/2019,"ImageNet,COCO",,Unreleased,Open (non-commercial),,None,2019
420,ALBERT,"Toyota Technological Institute at Chicago,Google Research",9/26/2019,"BookCorpus (BooksCorpus, Toronto Book Corpus),Wikipedia",,Open weights (unrestricted),Open source,,Indirect,2019
421,Adaptive Inputs + LayerDrop,"Facebook AI Research,LORIA",9/25/2019,WikiText-103,,Open weights (unrestricted),Open source,,Indirect,2019
416,T5-3B,Google,10/23/2019,C4,Confident,Open weights (unrestricted),Open source,"Third-party estimation,Reported",Direct,2019
417,M4-50B,Google,10/11/2019,,Confident,Unreleased,Unreleased,,Indirect,2019
422,Megatron-LM (8.3B),NVIDIA,9/17/2019,,Likely,Unreleased,Open source,"Hardware,Operation counting,Third-party estimation",None,2019
426,"Mogrifier (d2, MoS2, MC) + dynamic eval","DeepMind,University of Oxford",9/4/2019,WikiText-2,,Unreleased,Unreleased,,None,2019
424,ResNet-152 + ObjectNet,Massachusetts Institute of Technology (MIT),9/6/2019,ObjectNet,,Unreleased,Unreleased,Hardware,None,2019
425,UDSMProt,Fraunhofer Heinrich Hertz Institute,9/4/2019,"SwissProt,a subset of UniProtKB",Likely,Open weights (unrestricted),Open source,Operation counting,Indirect,2019
427,EN^2AS with performance reward,"Beijing Institute of Technology,University of Technology Sydney,Monash University",7/22/2019,,,Unreleased,Unreleased,,None,2019
428,Pluribus,Facebook AI Research,7/11/2019,,,Unreleased,Unreleased,Hardware,None,2019
415,T5-11B,Google,10/23/2019,C4,Confident,Open weights (unrestricted),Open source,"Reported,Operation counting,Third-party estimation",Direct,2019
429,BigBiGAN,Google,7/4/2019,ImageNet,,Open weights (unrestricted),Unreleased,,Indirect,2019
423,Megatron-BERT,NVIDIA,9/17/2019,,Confident,Unreleased,Open source,"Operation counting,Third-party estimation",Indirect,2019
414,BART-large,Facebook AI,10/29/2019,Wikipedia,,Open weights (unrestricted),Open source,,Indirect,2019
402,StarGAN v2,"NAVER,Yonsei University,Swiss Federal Institute of Technology",12/4/2019,"CelebA,AFHQ",Unknown,Open weights (non-commercial),Open (non-commercial),,Indirect,2019
412,Base LM + kNN LM + Continuous Cache,"Stanford University,Facebook AI Research",11/1/2019,WikiText-103,,Unreleased,Open source,,None,2019
430,RoBERTa Large,"Facebook,University of Washington",7/1/2019,"CC-News,BookCorpus (BooksCorpus, Toronto Book Corpus),WebText2,Wikipedia",Confident,Open weights (unrestricted),Open source,"Hardware,Operation counting",Indirect,2019
397,Big Transfer (BiT-L),Google Brain,12/24/2019,JFT-300M,,Unreleased,Unreleased,,None,2019
398,DD-PPO,"Georgia Institute of Technology,Facebook AI Research,Oregon State University,Simon Fraser University",12/19/2019,,Likely,Unreleased,Unreleased,Hardware,None,2019
399,OpenAI Five Rerun,OpenAI,12/13/2019,,,Unreleased,Unreleased,Third-party estimation,None,2019
400,OpenAI Five,OpenAI,12/13/2019,,Confident,Unreleased,Unreleased,,Indirect,2019
401,MMLSTM,"Beijing University of Posts and Telecommunications,University of West London",12/5/2019,WikiText-103,,Unreleased,Unreleased,,None,2019
403,Transformer-XL DeFINE (141M),"University of Washington,Allen Institute for AI",11/27/2019,"WikiText-103,Penn TreeBank",,Unreleased,Unreleased,,None,2019
404,Photo-Geometric Autoencoder,University of Oxford,11/25/2019,"CelebA,3DFAW,BFM",Unknown,Open weights (unrestricted),Open source,,Indirect,2019
405,Transformer - LibriVox + Decoding/Rescoring,Facebook,11/19/2019,"LibriSpeech,LibriVox",Confident,Open weights (unrestricted),,,Indirect,2019
406,MuZero,DeepMind,11/19/2019,,,Unreleased,Unreleased,Hardware,None,2019
407,MoCo,Facebook AI,11/13/2019,"ImageNet,Instagram-1B",,Open weights (non-commercial),Open (non-commercial),,Indirect,2019
408,Noisy Student (L2),"Carnegie Mellon University (CMU),Google",11/11/2019,"ImageNet,JFT",,Unreleased,Open source,Hardware,None,2019
409,Sandwich Transformer,"Allen Institute for AI,Facebook AI Research",11/10/2019,"BookCorpus (BooksCorpus, Toronto Book Corpus),enwik8,text8",,Unreleased,Open (non-commercial),,None,2019
410,CamemBERT,"Facebook,INRIA,Sorbonne University",11/10/2019,CCNet,Confident,Open weights (unrestricted),Unreleased,"Hardware,Operation counting",Indirect,2019
411,XLM-RoBERTa,Facebook AI,11/5/2019,CC100,Confident,Open weights (non-commercial),Open (non-commercial),Operation counting,Indirect,2019
413,AlphaStar,DeepMind,10/30/2019,,,Unreleased,Open source,Hardware,None,2019
431,Tensorized Transformer (257M),"Tianjin University,Microsoft Research Asia,Beijing Institute of Technology",6/24/2019,WikiText-103,,Unreleased,Open (non-commercial),,None,2019
454,Transformer-XL + RMS dynamic eval,University of Edinburgh,4/17/2019,WikiText-103,,Unreleased,Open source,,None,2019
433,LaNet-L (CIFAR-10),"Brown University,Facebook",6/17/2019,CIFAR-10,Confident,Open weights (non-commercial),Open (non-commercial),,Indirect,2019
453,SpecAugment,Google Brain,4/18/2019,"LibriSpeech,Switchboard,Fisher",Unknown,Unreleased,Unreleased,,None,2019
455,WeNet (Penn Treebank),Amazon,4/8/2019,Penn TreeBank,Likely,Unreleased,Unreleased,"Hardware,Operation counting",None,2019
456,True-Regularization+Finetune+Dynamic-Eval,"Mobvoi,Williams College",4/8/2019,Penn TreeBank,,Unreleased,Unreleased,,None,2019
457,Cross-lingual alignment,"Tel Aviv University,Massachusetts Institute of Technology (MIT)",4/4/2019,"Wikipedia,CoNLL2017",,Open weights (unrestricted),Open source,Hardware,Indirect,2019
458,FAIRSEQ Adaptive Inputs,"Facebook AI Research,Google Brain",4/1/2019,WikiText-103,,Unreleased,Open source,,None,2019
459,SciBERT,Allen Institute for AI,3/26/2019,,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2019
452,BERT-Large-CAS (PTB+WT2+WT103),Amazon,4/20/2019,"Penn TreeBank,WikiText-2,WikiText-103",,Unreleased,Open source,,None,2019
432,Walking Minotaur robot,"University of California (UC) Berkeley,Google Brain",6/19/2019,,Unknown,Unreleased,Unreleased,,None,2019
463,GPT-2 (1.5B),OpenAI,2/14/2019,WebText,,Open weights (unrestricted),Unreleased,Operation counting,Direct,2019
464,Hanabi 4 player,"DeepMind,University of Oxford,Carnegie Mellon University (CMU),Google Brain",2/1/2019,,,Unreleased,Unreleased,Hardware,None,2019
465,MT-DNN,Microsoft,1/31/2019,"GLUE,SciTail",,Open weights (unrestricted),Open source,,Indirect,2019
466,Transformer-XL (257M),"Carnegie Mellon University (CMU),Google Brain",1/9/2019,WikiText-103,,Open weights (unrestricted),Open source,,Indirect,2019
467,Decoupled weight decay regularization,University of Freiburg,1/4/2019,CIFAR-10,,Open weights (unrestricted),Open source,Operation counting,Indirect,2019
468,Transformer ELMo,"Allen Institute for AI,University of Washington",1/1/2019,,,Unreleased,Unreleased,,None,2019
461,KataGo,Jane Street,2/27/2019,,Speculative,Open weights (unrestricted),Open source,Hardware,Indirect,2019
451,DANet,Chinese Academy of Sciences,4/21/2019,"Cityscapes,COCO-Stuff,PASCAL-Context",Unknown,Open weights (unrestricted),Open source,,Indirect,2019
460,NMT Transformer 437M,"Google,Bar-Ilan University",2/28/2019,,Confident,Unreleased,Unreleased,,Indirect,2019
449,ResNet-50 Billion-scale,Facebook AI,5/2/2019,"YFCC-100M,IG-1B-Targeted",,Open weights (non-commercial),Unreleased,,Indirect,2019
450,Neuro-Symbolic Concept Learner,"Massachusetts Institute of Technology (MIT),Tsinghua University,MIT-IBM Watson AI Lab,DeepMind",4/26/2019,"CLEVR,VQS,ImageNet",Unknown,Unreleased,Open source,,None,2019
434,PG-SWGAN,ETH Zurich,6/15/2019,"CIFAR-10,LSUN,CelebA",Unknown,Unreleased,Open (non-commercial),,None,2019
435,FixRes ResNeXt-101 WSL,Facebook AI,6/14/2019,ImageNet,,Open weights (non-commercial),Open (non-commercial),,Indirect,2019
436,Char-CNN-BiLSTM,Capital One,6/13/2019,,Unknown,Unreleased,Unreleased,,None,2019
437,AWD-LSTM + MoS + Partial Shuffled,University of Texas at Austin,6/10/2019,WikiText-2,,Open weights (non-commercial),Open (non-commercial),,Indirect,2019
438,Transformer-XL Large + Phrase Induction,"Massachusetts Institute of Technology (MIT),University of Illinois Urbana-Champaign (UIUC)",6/4/2019,WikiText-103,,Unreleased,Open source,,None,2019
439,AMDIM,Microsoft Research,6/3/2019,"ImageNet,CIFAR-10",,Open weights (unrestricted),Open source,,Indirect,2019
440,XLNet,"Carnegie Mellon University (CMU),Google Brain",6/1/2019,"Wikipedia,BookCorpus (BooksCorpus, Toronto Book Corpus)",Confident,Open weights (unrestricted),Open source,"Hardware,Operation counting",Indirect,2019
462,ProxylessNAS,Massachusetts Institute of Technology (MIT),2/23/2019,ImageNet,,Open weights (unrestricted),Open source,Hardware,Indirect,2019
442,DLRM-2020,Facebook AI,5/31/2019,,,Unreleased,Open source,Reported,Indirect,2019
441,XLM,Facebook,6/1/2019,,,Open weights (non-commercial),Open (non-commercial),,Indirect,2019
447,AWD-LSTM-DRILL + dynamic evaluation† (WT2),IDIAP,5/14/2019,WikiText-2,,Open weights (unrestricted),Open (restricted use),,Indirect,2019
446,CPC v2,"DeepMind,University of California (UC) Berkeley",5/22/2019,ImageNet,,Unreleased,Unreleased,,None,2019
448,ResNeXt-101 Billion-scale,Facebook AI,5/2/2019,YFCC-100M,,Open weights (non-commercial),Unreleased,,Indirect,2019
444,MnasNet-A1 + SSDLite,Google,5/29/2019,COCO,Speculative,Open weights (unrestricted),Open source,Hardware,Indirect,2019
443,MnasNet-A3,Google,5/29/2019,ImageNet,Speculative,Open weights (unrestricted),Open source,Hardware,Indirect,2019
445,EfficientNet-L2,Google,5/28/2019,ImageNet,,Open weights (unrestricted),Open source,,Indirect,2019
378,Go-explore,"Uber AI,OpenAI",4/27/2020,,Unknown,Unreleased,Open (non-commercial),,None,2020
379,CURL,University of California (UC) Berkeley,4/8/2020,,,Open weights (unrestricted),Open source,,Indirect,2020
385,TransformerXL + spectrum control,"University of California Los Angeles (UCLA),JD.com",3/11/2020,WikiText-103,,Unreleased,Unreleased,,None,2020
380,Agent57,DeepMind,3/30/2020,,Unknown,Unreleased,Unreleased,,None,2020
381,MetNet,Google,3/24/2020,,Unknown,Unreleased,Unreleased,,None,2020
382,ELECTRA,"Stanford University,Google,Google Brain",3/23/2020,"BookCorpus (BooksCorpus, Toronto Book Corpus),Wikipedia,ClueWeb,Gigaword",,Open weights (unrestricted),Open source,Reported,Indirect,2020
383,Tensor-Transformer(1core)+PN (WT103),University of California (UC) Berkeley,3/17/2020,WikiText-103,,Open weights (unrestricted),Open source,,Indirect,2020
384,Routing Transformer (WT-103),Google Research,3/12/2020,WikiText-103,,Open weights (unrestricted),Unreleased,,Indirect,2020
386,TCAN (WT2),"Nanjing University,Ant Group",2/28/2020,WikiText-2,,Unreleased,Open source,,None,2020
390,ALBERT-xxlarge,"Toyota Technological Institute at Chicago,Google",2/9/2020,"Wikipedia,BookCorpus (BooksCorpus, Toronto Book Corpus)",,Open weights (unrestricted),Open source,Hardware,Indirect,2020
388,Turing-NLG,Microsoft,2/13/2020,,Likely,Unreleased,Unreleased,"Third-party estimation,Operation counting",None,2020
389,SimCLR,Google Brain,2/13/2020,ILSVRC 2012 subset of ImageNet,,Open weights (unrestricted),Open source,,Indirect,2020
391,TaLK Convolution,Carleton University,2/8/2020,WikiText-103,,Unreleased,Unreleased,,None,2020
392,Perceiver IO (optical flow),DeepMind,2/8/2020,AutoFlow,,Unreleased,Unreleased,,None,2020
393,Theseus 6/768,"University of California San Diego,Beihang University,Microsoft",2/7/2020,GLUE,,Open weights (unrestricted),Open source,,Indirect,2020
394,Meena,Google Brain,1/28/2020,,Confident,Unreleased,Unreleased,"Hardware,Operation counting,Third-party estimation",Direct,2020
396,AlphaFold,DeepMind,1/15/2020,"PDB (Protein Data Bank),UniRef30 (FKA UniClust30)",Speculative,Unreleased,Unreleased,"Hardware,Third-party estimation",None,2020
377,Once for All,"MIT-IBM Watson AI Lab,Massachusetts Institute of Technology (MIT),IBM",4/29/2020,ImageNet,,Open weights (unrestricted),Open source,Hardware,Indirect,2020
387,Feedback Transformer,"LORIA,University of Lorraine,Facebook AI Research",2/21/2020,WikiText-103,,Unreleased,Unreleased,,None,2020
395,ContextNet + Noisy Student,Google,1/19/2020,"LibriSpeech,LibriLight",Confident,Unreleased,Unreleased,Hardware,Indirect,2020
376,ATLAS,"Allen Institute for AI,University of Washington",5/2/2020,SQuAD 1.1,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2020
374,NAS+ESS (156M),"Northeastern University (China),Chinese Academy of Sciences,NiuTrans Research,Kingsoft",5/6/2020,Penn TreeBank,,Unreleased,Unreleased,,None,2020
375,UnifiedQA,"Allen Institute for AI,University of Washington",5/2/2020,,Confident,Unreleased,,"Operation counting,Hardware",Indirect,2020
343,ERNIE-Doc (247M),Baidu,12/31/2020,WikiText-103,,Open weights (unrestricted),Unreleased,,Indirect,2020
344,CT-MoS (WT2),"Google,National Tsing Hua University",12/25/2020,WikiText-2,,Unreleased,Unreleased,,None,2020
345,DensePhrases,"Korea University,Princeton University",12/23/2020,"SQuAD,NQ (Natural Questions)",Speculative,Open weights (unrestricted),Open source,Hardware,Indirect,2020
346,VQGAN + CLIP,Heidelberg University,12/17/2020,,Unknown,,,,None,2020
347,ESM1b,"Facebook AI Research,New York University (NYU)",12/15/2020,UniRef50,Confident,Open weights (unrestricted),Unreleased,"Hardware,Operation counting",Indirect,2020
348,CPM-Large,"Tsinghua University,Beijing Academy of Artificial Intelligence / BAAI",12/1/2020,Unspecified unreleased,,Open weights (unrestricted),Unreleased,Third-party estimation,Indirect,2020
349,AlphaFold 2,DeepMind,11/30/2020,"PDB (Protein Data Bank),UniRef30 (FKA UniClust30),UniRef90,MGnify,BFD (Big Fantastic Dataset),UniProtKB",Likely,Open weights (unrestricted),Unreleased,Hardware,Indirect,2020
351,SimCLRv2,Google Brain,10/26/2020,,,,,,None,2020
352,wave2vec 2.0 LARGE,Facebook,10/22/2020,"LibriSpeech,LibriLight",,Open weights (unrestricted),Open source,Hardware,Indirect,2020
353,ViT-Huge/14,"Google Brain,Google Research",10/22/2020,"ImageNet-1k,ImageNet21k,JFT-300M",Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2020
354,ViT-Base/32,Google Brain,10/22/2020,JFT-300M,,,,,None,2020
355,German ELECTRA Large,"deepset,Bayerische Staatsbibliothek Muenchen",10/21/2020,"Wikipedia,OPUS,OSCAR,OpenLegalData",Confident,Open weights (unrestricted),,"Hardware,Operation counting",Indirect,2020
356,GBERT-Large,"deepset,Bayerische Staatsbibliothek Muenchen",10/21/2020,"Wikipedia,OPUS,OSCAR,OpenLegalData",Likely,Open weights (unrestricted),Unreleased,Hardware,Indirect,2020
357,mT5-XXL,"Google,Google Research",10/20/2020,mC4,Confident,Open weights (unrestricted),Open source,Operation counting,Direct,2020
350,KEPLER,"Tsinghua University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),HEC,CIFAR AI Research,Princeton University,University of Montreal / Université de Montréal",11/23/2020,"Wikipedia,BookCorpus (BooksCorpus, Toronto Book Corpus),Wikidata5M",,Unreleased,Open source,Hardware,None,2020
359,LUKE,"University of Washington,National Institute of Informatics",10/2/2020,Wikipedia,Likely,Open weights (unrestricted),Open source,Hardware,Indirect,2020
358,Conformer + Wav2vec 2.0 + Noisy Student,"Google,Google Research,Google Brain",10/20/2020,LibriLight,Confident,Unreleased,Unreleased,Hardware,Indirect,2020
373,ContextNet,Google,5/7/2020,LibriSpeech,Likely,Unreleased,Unreleased,,None,2020
372,Conformer,Google,5/16/2020,LibriSpeech,Confident,Unreleased,Unreleased,,Indirect,2020
371,Retrieval-Augmented Generator,"Facebook,New York University (NYU),University College London (UCL)",5/22/2020,"Wikipedia,NQ (Natural Questions)",Confident,Open weights (unrestricted),Unreleased,,Indirect,2020
370,DETR,Facebook,5/26/2020,COCO 2017,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2020
368,iGPT-L,OpenAI,6/17/2020,ILSVRC 2012 subset of ImageNet,,Open weights (unrestricted),Open source,Hardware,Indirect,2020
367,iGPT-XL,OpenAI,6/17/2020,ILSVRC 2012 subset of ImageNet,,Open weights (unrestricted),Open source,Third-party estimation,Indirect,2020
369,GPT-3 175B (davinci),OpenAI,5/28/2020,"Common Crawl,WebText2,Wikipedia,Books1,Books2",Confident,API access,Unreleased,Reported,Direct,2020
365,SemExp,"Carnegie Mellon University (CMU),Facebook AI Research",7/2/2020,"Gibson,Matterport3D (MP3D)",Unknown,Open weights (unrestricted),Open source,,Indirect,2020
364,Hopfield Networks (2020),"Johannes Kepler University Linz,Institute of Advanced Research in Artificial Intelligence,University of Oslo",7/16/2020,"BACE,SIDER",Unknown,Open weights (unrestricted),Unreleased,,Indirect,2020
363,EfficientDet,Google Brain,7/27/2020,COCO 2017,,Open weights (unrestricted),Open source,,Indirect,2020
362,DeLighT,"University of Washington,Allen Institute for AI,Facebook AI Research",8/3/2020,WikiText-103,,Unreleased,Open source,,None,2020
361,ERNIE-GEN (large),Baidu,8/6/2020,"CC-News,BookCorpus (BooksCorpus, Toronto Book Corpus),WebText2,Wikipedia,C4",Speculative,Open weights (non-commercial),Open (non-commercial),Operation counting,Indirect,2020
360,ProBERTa,"University of Illinois Urbana-Champaign (UIUC),Reed College",9/1/2020,UniProtKB/Swiss-Prot,Confident,,,Hardware,Indirect,2020
366,GShard (dense),Google,6/30/2020,,Confident,Unreleased,Open source,"Operation counting,Hardware",Indirect,2020
287,EfficientZero,"Tsinghua University,University of California (UC) Berkeley,Shanghai Qi Zhi institute",10/30/2021,,Unknown,,,,None,2021
292,Megatron-Turing NLG 530B,"Microsoft,NVIDIA",10/11/2021,"Common Crawl,The Pile,CC-Stories,Realnews",,Unreleased,Unreleased,Third-party estimation,None,2021
288,Eve,"Harvard Medical School,University of Oxford",10/27/2021,UniRef100,Likely,Unreleased,Open source,,None,2021
289,base LM+GNN+kNN,"Shannon.AI,Nanjing University,Nanyang Technological University,Zhejiang University",10/17/2021,WikiText-103,,Open weights (unrestricted),Open source,,Indirect,2021
290,T0-XXL,"Hugging Face,Brown University",10/15/2021,P3 (Public Pool of Prompts),Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2021
291,Yuan 1.0,Inspur,10/12/2021,"Common Crawl,Wikipedia,Sogue News",Confident,API access,Unreleased,Reported,Indirect,2021
293,AlphaFold-Multimer,"Google DeepMind,DeepMind",10/4/2021,PDB (Protein Data Bank),Confident,Open weights (unrestricted),Unreleased,Hardware,Indirect,2021
302,Zidong Taichu,"Chinese Academy of Sciences,Wuhan AI Computing Center",8/11/2021,,Confident,,,Operation counting,Indirect,2021
295,PLATO-XL,Baidu,9/20/2021,,Confident,Open weights (unrestricted),,Operation counting,Indirect,2021
296,HyperCLOVA 204B,NAVER,9/10/2021,Unspecified unreleased,Speculative,,Unreleased,,None,2021
297,PermuteFormer,Peking University,9/6/2021,WikiText-103,Speculative,Unreleased,Open source,Operation counting,None,2021
298,MEB,Microsoft,9/4/2021,,,,,,None,2021
299,FLAN 137B,Google Research,9/3/2021,"Wikipedia,Unspecified unreleased",Confident,Unreleased,Unreleased,Operation counting,Indirect,2021
301,DNABERT,Northeastern University,8/15/2021,Human Reference Genome (GRCh38/hg38),Confident,Open weights (unrestricted),Open source,"Hardware,Operation counting",Indirect,2021
286,S4,Stanford University,10/31/2021,WikiText-103,Likely,Open weights (unrestricted),Open source,,Indirect,2021
294,TrOCR,"Beihang University,Microsoft Research Asia",9/21/2021,,Confident,Open weights (unrestricted),Open source,,Indirect,2021
285,CodeT5-base,"Salesforce,Nanyang Technological University",11/1/2021,"CodeSearchNet,BigQuery",Likely,Open weights (unrestricted),Open source,Hardware,Direct,2021
269,ERNIE 3.0 Titan,"Baidu,Peng Cheng Laboratory",12/23/2021,ERNIE 3.0 Corpus,Confident,Hosted access (no API),Unreleased,Operation counting,Indirect,2021
283,Masked Autoencoders ViT-H,Facebook AI Research,11/11/2021,ImageNet-1k,Speculative,Open weights (non-commercial),Open (non-commercial),"Hardware,Operation counting",Indirect,2021
268,ERNIE-ViLG,Baidu,12/31/2021,,,,,,None,2021
303,Jurassic-1-Jumbo,AI21 Labs,8/11/2021,,,API access,Unreleased,Third-party estimation,None,2021
270,XGLM-7.5B,"Meta AI,Facebook AI Research",12/20/2021,"Subset of CC100-XL,CC100-XL,Common Crawl",Confident,Open weights (non-commercial),Unreleased,"Operation counting,Hardware",Indirect,2021
271,LDM-1.45B,"Heidelberg University,Runway",12/20/2021,LAION-400M,Confident,Open weights (unrestricted),Open source,,Indirect,2021
272,GLIDE,OpenAI,12/20/2021,DALL-E,Speculative,,,Comparison with other models,None,2021
273,Contriever,"Meta AI,University College London (UCL),PSL University,Université Grenoble Alpes",12/16/2021,"Wikipedia,CCNet",Likely,Open weights (non-commercial),Open (non-commercial),Operation counting,Indirect,2021
274,LongT5,Google Research,12/15/2021,C4,Confident,Open weights (unrestricted),Open source,,Direct,2021
275,GLaM,Google,12/13/2021,"Wikipedia,GLaM dataset",Confident,Unreleased,Unreleased,"Operation counting,Hardware",Indirect,2021
276,Gopher (280B),DeepMind,12/8/2021,MassiveTex,Confident,Unreleased,Unreleased,Reported,Indirect,2021
277,Student of Games,DeepMind,12/6/2021,,Speculative,Unreleased,Unreleased,,None,2021
278,N√úWA,"Microsoft Research,Peking University",11/24/2021,"Conceptual Captions (CC3M),Moments in Time,VATEX",,Unreleased,Unreleased,Hardware,None,2021
279,Florence,Microsoft,11/22/2021,FLD-900M,Confident,Unreleased,Unreleased,Hardware,Indirect,2021
280,BASIC-L,Google,11/19/2021,"JFT,ALIGN",Likely,Unreleased,Unreleased,Hardware,None,2021
281,Swin Transformer V2 (SwinV2-G),Microsoft Research Asia,11/18/2021,ImageNet21k,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2021
282,ViT-G/14 (LiT),Google,11/15/2021,"Conceptual Captions 12M (CC12M),YFCC-100M,Unspecified unreleased",Confident,,,,Indirect,2021
284,Projected GAN,Heidelberg University,11/1/2021,,Confident,,,Hardware,Indirect,2021
304,W2v-BERT,"Google Brain,Massachusetts Institute of Technology (MIT)",8/7/2021,LibriLight,Confident,,,,Indirect,2021
300,XLMR-XXL,Facebook AI Research,8/17/2021,CC100,Confident,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2021
306,6-Act Tether,"Facebook AI Research,Georgia Institute of Technology",8/3/2021,Matterport,Confident,,,,Indirect,2021
327,ProtBERT-BFD,"Technical University of Munich,NVIDIA,Seoul National University,Google,Oak Ridge National Laboratory,Med AI Technology",5/4/2021,BFD (Big Fantastic Dataset),Confident,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2021
328,ViT + DINO,"INRIA,Facebook AI Research",4/29/2021,ImageNet,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2021
329,PLUG,Alibaba,4/19/2021,,,Hosted access (no API),Unreleased,Hardware,None,2021
330,M6-T,Alibaba,3/5/2021,M6-Corpus,Likely,Unreleased,Unreleased,Third-party estimation,None,2021
331,Generative BST,Facebook AI Research,3/5/2021,,Confident,Open weights (unrestricted),,Operation counting,Indirect,2021
332,Meta Pseudo Labels,"Google Brain,Google AI",3/1/2021,"ImageNet,JFT-300M",,Unreleased,Open source,Hardware,None,2021
334,Rational DQN Average,TU Darmstadt,2/18/2021,,,,,,None,2021
326,ProtT5-XXL,"Technical University of Munich,Med AI Technology,NVIDIA,Oak Ridge National Laboratory,Google,Seoul National University",5/4/2021,"BFD (Big Fantastic Dataset),UniRef50",Confident,Open weights (unrestricted),Unreleased,"Third-party estimation,Operation counting",Direct,2021
335,MSA Transformer,"Facebook AI Research,University of California (UC) Berkeley,New York University (NYU)",2/13/2021,"UniRef50,UniRef30 (FKA UniClust30)",Likely,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2021
337,DeiT-B,"Meta AI,Sorbonne University",1/15/2021,ImageNet,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2021
338,Switch,Google,1/11/2021,C4,,Open weights (unrestricted),Unreleased,Third-party estimation,Indirect,2021
339,BigSSL,"Google,Apple",1/10/2021,,,,,,None,2021
340,DALL-E,OpenAI,1/5/2021,DALL-E,,API access,Unreleased,Third-party estimation,None,2021
341,CLIP (ViT L/14@336px),OpenAI,1/5/2021,Unspecified unreleased,,Open weights (unrestricted),Unreleased,Third-party estimation,Indirect,2021
305,YOLOX-X,Megvii Inc,8/6/2021,COCO 2017,Likely,Open weights (unrestricted),Open source,Operation counting,Indirect,2021
342,CLIP (ResNet-50),OpenAI,1/5/2021,,,,,,None,2021
336,top-down frozen classifier,"University of Edinburgh,Toshiba Cambridge Research Laboratory",2/9/2021,WSJ,Unknown,Unreleased,Unreleased,,None,2021
325,ProtT5-XXL-BFD,"Technical University of Munich,Med AI Technology,NVIDIA,Oak Ridge National Laboratory,Google,Seoul National University",5/4/2021,BFD (Big Fantastic Dataset),Confident,Open weights (unrestricted),Unreleased,Operation counting,Direct,2021
333,SRU++ Large,ASAPP,2/24/2021,enwik8,,Open weights (unrestricted),Open source,,Indirect,2021
323,MedBERT,"Peng Cheng Laboratory,University of Texas at Houston",5/20/2021,Cerner Health Facts,Likely,Unreleased,Open source,Hardware,None,2021
307,SEER,"Facebook AI Research,INRIA",7/29/2021,Instagram,,Open weights (non-commercial),Open (non-commercial),Hardware,Indirect,2021
308,HuBERT,Facebook AI Research,7/27/2021,"LibriSpeech,LibriLight",Speculative,Open weights (unrestricted),Open source,Hardware,Indirect,2021
324,ADM,OpenAI,5/11/2021,"LSUN,ILSVRC 2012 subset of ImageNet",Confident,Open weights (non-commercial),Open source,Hardware,Indirect,2021
309,GOAT,DeepMind,7/27/2021,XLand,Speculative,Unreleased,Unreleased,Hardware,None,2021
310,Codex,OpenAI,7/7/2021,,Likely,API access,Unreleased,,None,2021
312,Adaptive Input Transformer + RD,"Microsoft Research Asia,Soochow University",6/28/2021,WMT14,,Unreleased,Open source,,None,2021
313,EfficientNetV2-XL,"Google,Google Brain",6/23/2021,"ImageNet21k,ILSVRC 2012 subset of ImageNet",Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2021
314,Denoising Diffusion Probabilistic Models (LSUN Bedroom),University of California (UC) Berkeley,6/11/2021,LSUN Bedroom,,Open weights (unrestricted),Open source,Hardware,Indirect,2021
311,ERNIE 3.0,Baidu,7/5/2021,,,Open weights (unrestricted),Open source,Operation counting,Indirect,2021
320,ByT5-XXL,"Google,Google Research",5/28/2021,mC4,Likely,Open weights (unrestricted),Open source,Operation counting,Direct,2021
316,DeBERTa,Microsoft,6/10/2021,"Wikipedia,CC-Stories,OPENWEBTEXT,BookCorpus (BooksCorpus, Toronto Book Corpus)",,Open weights (unrestricted),Open source,Hardware,Indirect,2021
317,EMDR,"Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),McGill University,DeepMind",6/9/2021,"Wikipedia,NQ (Natural Questions),TriviaQA",Confident,Open weights (unrestricted),Open source,"Hardware,Operation counting",Indirect,2021
318,CoAtNet,"Google,Google Research,Google Brain",6/9/2021,JFT-3B,Confident,Unreleased,Unreleased,Hardware,Indirect,2021
319,ViT-G/14,"Google Brain,Google Research",6/8/2021,"JFT-3B,ImageNet",Confident,Unreleased,Open source,"Hardware,Operation counting",Indirect,2021
322,CogView,"Tsinghua University,Alibaba DAMO Academy",5/26/2021,WuDao Corpora,Likely,Open weights (unrestricted),Open source,Third-party estimation,Indirect,2021
315,ALIGN,Google Research,6/11/2021,"Conceptual Captions (CC3M),FIT400M",Confident,Unreleased,Unreleased,Hardware,Indirect,2021
321,Transformer local-attention (NesT-B),"Google Cloud,Google Research",5/26/2021,ImageNet-1k,,Open weights (unrestricted),Open source,Operation counting,Indirect,2021
211,DiffDock,Massachusetts Institute of Technology (MIT),10/4/2022,PDB (Protein Data Bank),Likely,Open weights (unrestricted),,Hardware,Indirect,2022
210,Phenaki,"University College London (UCL),University of Michigan,Google Brain",10/5/2022,"LAION-400M,Unspecified unreleased",,,,,None,2022
209,Diplodocus,"Meta AI,Massachusetts Institute of Technology (MIT)",10/11/2022,,Unknown,Open weights (non-commercial),Open source,,Indirect,2022
205,LMSI-Palm,"Google,University of Illinois Urbana-Champaign (UIUC)",10/20/2022,GSM8K,Confident,Unreleased,,,Direct,2022
207,Flan-PaLM 540B,Google,10/20/2022,Flan,Confident,Unreleased,Unreleased,"Reported,Hardware",Direct,2022
206,Flan-T5 11B,Google,10/20/2022,,Confident,Open weights (unrestricted),Unreleased,Reported,Direct,2022
212,Make-A-Video,Meta AI,9/29/2022,"LAION,WebVid-10M,HD-VILA-100M",Unknown,,,,None,2022
208,GenSLM,"University of Chicago,NVIDIA,Harvard University,Cerebras Systems,Technical University of Munich,California Institute of Technology",10/11/2022,"SARS-CoV-2 genome dataset,BV-BRC",Confident,,,Reported,Indirect,2022
213,Whisper,OpenAI,9/21/2022,Unspecified unreleased,Likely,Open weights (unrestricted),Unreleased,Hardware,Indirect,2022
220,ESM2-15B,"Meta AI,New York University (NYU),Stanford University,Massachusetts Institute of Technology (MIT)",7/21/2022,UniRef50,Confident,Open weights (unrestricted),Unreleased,"Hardware,Third-party estimation",Indirect,2022
215,BEIT-3,Microsoft,8/22/2022,"ImageNet21k,COCO,English Wikipedia,BookCorpus (BooksCorpus, Toronto Book Corpus)",Likely,Unreleased,,Operation counting,None,2022
216,BlenderBot 3,"McGill University,Meta AI,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms)",8/10/2022,BlenderBot 3 Data,Likely,Open weights (non-commercial),Open source,Operation counting,Indirect,2022
217,GLM-130B,Tsinghua University,8/4/2022,"The Pile,WuDao Corpora",Confident,Open weights (non-commercial),Unreleased,"Operation counting,Hardware",Indirect,2022
218,AlexaTM 20B,Amazon,8/2/2022,"mC4,Wikipedia",Confident,API access,,Hardware,Indirect,2022
219,OmegaPLM,"Massachusetts Institute of Technology (MIT),Westlake University",7/22/2022,UniRef50,Confident,,,Hardware,Indirect,2022
221,BLOOM-176B,"Hugging Face,BigScience",7/11/2022,BigScience ROOTS Corpus,Confident,Open weights (restricted use),Unreleased,Hardware,Direct,2022
222,NLLB,Meta AI,7/6/2022,,,Open weights (unrestricted),Open source,Hardware,Indirect,2022
204,U-PaLM (540B),Google,10/20/2022,,Confident,Unreleased,Unreleased,Comparison with other models,Direct,2022
214,PaLI,Google,9/14/2022,WebLI,Likely,Unreleased,Unreleased,"Operation counting,Hardware",None,2022
203,EnCodec,Meta AI,10/24/2022,"DNS,Common Voice,AudioSet,FSD50K,Jamendo",Unknown,Open weights (non-commercial),Open source,,Indirect,2022
184,RT-1,Google,12/13/2022,RT-1,Confident,Open weights (unrestricted),Open source,,Indirect,2022
201,BLOOMZ-176B,Hugging Face,11/3/2022,xP3,Likely,Open weights (unrestricted),Open source,,Direct,2022
223,CodeT5-large,Salesforce,7/5/2022,GitHub,Likely,Open weights (unrestricted),,Hardware,Direct,2022
182,Hybrid H3-2.7B,"Stanford University,University at Buffalo",12/28/2022,The Pile,,Open weights (unrestricted),Unreleased,,Indirect,2022
183,CaLM,University of Oxford,12/19/2022,European Nucleotide Archive (ENA),Likely,,,"Hardware,Operation counting",None,2022
185,TranceptEve,"University of Oxford,Harvard Medical School",12/10/2022,ProteinGym,Unknown,,,,None,2022
186,DeepNash,DeepMind,12/1/2022,,Unknown,,,,None,2022
188,GPT-3.5,OpenAI,11/28/2022,,Speculative,API access,Unreleased,"Comparison with other models,Benchmarks",None,2022
189,DiT-XL/2 + Discriminator Guidance,"Korea Advanced Institute of Science and Technology (KAIST),NAVER",11/28/2022,,Unknown,,,,None,2022
190,Discriminator Guidance,"Korea Advanced Institute of Science and Technology (KAIST),NAVER",11/28/2022,,Confident,Open weights (non-commercial),Open (non-commercial),Hardware,Indirect,2022
202,eDiff-I,NVIDIA,11/2/2022,Unspecified unreleased,Likely,API access,,Operation counting,None,2022
191,ALM 1.0,Beijing Academy of Artificial Intelligence / BAAI,11/28/2022,ArabicText 2022,Speculative,,,,None,2022
193,AR-LDM,"Alibaba,University of Waterloo,Vector Institute",11/20/2022,,Confident,Unreleased,Open (non-commercial),Hardware,Indirect,2022
194,Fusion in Encoder,Samsung,11/18/2022,TriviaQA,Likely,,,Hardware,None,2022
195,Galactica,Meta AI,11/16/2022,Galactica Corpus,Likely,Open weights (non-commercial),Unreleased,Operation counting,Indirect,2022
196,EVA-01,"Beijing Academy of Artificial Intelligence / BAAI,Huazhong University of Science and Technology,Zhejiang University,Beijing Institute of Technology",11/14/2022,"ImageNet21k,COCO,Conceptual Captions 12M (CC12M),Conceptual Captions (CC3M)",Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2022
197,AltCLIP_M9,Beijing Academy of Artificial Intelligence / BAAI,11/12/2022,"Conceptual Captions (CC3M),LAION-400M,TSL2019,OPUS,WuDao Corpora,LAION-2B",Unknown,Open weights (unrestricted),Open source,,Indirect,2022
198,InternImage,"Shanghai AI Lab,Tsinghua University,Nanjing University,SenseTime,Chinese University of Hong Kong (CUHK)",11/10/2022,"LAION-400M,Conceptual Captions 12M (CC12M),ImageNet-1k",Confident,Open weights (unrestricted),,Operation counting,Indirect,2022
199,mT0-13B,"Hugging Face,BigScience",11/3/2022,xP3,Confident,Open weights (unrestricted),Unreleased,,Indirect,2022
200,Mogrifier RLSTM (WT2),DeepMind,11/3/2022,WikiText-2,Confident,Unreleased,Unreleased,Operation counting,Indirect,2022
192,CICERO,Meta AI,11/22/2022,WebDiplomacy,Unknown,Open weights (non-commercial),Open source,,Indirect,2022
224,Minerva (540B),Google,6/29/2022,arXiv,,Unreleased,Unreleased,Hardware,None,2022
187,GPT-3.5 Turbo,OpenAI,11/30/2022,Unspecified unreleased,Speculative,API access,Unreleased,,None,2022
226,Parti,Google Research,6/22/2022,"LAION-400M,FIT400M,JFT-4B",,Unreleased,Unreleased,Operation counting,None,2022
250,DeepNet,Microsoft Research,3/1/2022,"CCMatrix,OPUS",,,,,None,2022
251,PolyCoder,Carnegie Mellon University (CMU),2/26/2022,,Likely,,,Hardware,None,2022
252,ST-MoE,"Google,Google Brain,Google Research",2/17/2022,C4,Likely,Unreleased,Open source,Operation counting,None,2022
253,Midjourney V1,Midjourney,2/15/2022,Unspecified unreleased,Unknown,Hosted access (no API),Unreleased,,None,2022
254,ProteinBERT,"Hebrew University of Jerusalem,Ben-Gurion University of the Negev,Deep Trading",2/10/2022,UniRef90,Confident,,,Hardware,Indirect,2022
255,LaMDA,Google,2/10/2022,Infiniset,Confident,Unreleased,Unreleased,Hardware,Indirect,2022
256,GPT-NeoX-20B,EleutherAI,2/9/2022,The Pile,,Open weights (unrestricted),Open source,Hardware,Indirect,2022
257,RETRO-7B,DeepMind,2/7/2022,WikiText-103,,Unreleased,Unreleased,Operation counting,None,2022
249,Statement Curriculum Learning,OpenAI,3/2/2022,"Common Crawl,WebMath",,,,,None,2022
258,AlphaCode,DeepMind,2/2/2022,"CodeContests,Unspecified unreleased",,Unreleased,Unreleased,Hardware,None,2022
261,InstructGPT 1.3B,OpenAI,1/27/2022,,Confident,,,,Indirect,2022
262,OntoProtein,Zhejiang University,1/23/2022,ProteinKG25,,,,,None,2022
263,AbLang (heavy sequences),University of Oxford,1/22/2022,Observed Antibody Space (OAS) database,Confident,,,,Indirect,2022
264,data2vec (vision),Meta AI,1/20/2022,ImageNet-1k,,,,,None,2022
265,data2vec (speech),Meta AI,1/20/2022,LibriSpeech,,,,,None,2022
266,data2vec (language),Meta AI,1/20/2022,"BookCorpus (BooksCorpus, Toronto Book Corpus),English Wikipedia",,Open weights (unrestricted),Open source,,Indirect,2022
267,Detic,"Meta AI,University of Texas at Austin",1/7/2022,"ImageNet21k,Conceptual Captions (CC3M),LVIS",Speculative,Open weights (unrestricted),Open source,Hardware,Indirect,2022
225,ProGen2-xlarge,"Salesforce Research,Columbia University,Johns Hopkins University",6/27/2022,"UniRef90,BFD30",Confident,Open weights (unrestricted),Unreleased,"Hardware,Third-party estimation",Indirect,2022
260,InstructGPT 6B,OpenAI,1/27/2022,,Confident,,,,Indirect,2022
248,MegaSyn,Collaborations Pharmaceuticals,3/7/2022,ChEMBL,Unknown,Unreleased,,,None,2022
259,InstructGPT 175B,OpenAI,1/27/2022,,Confident,,,Reported,Indirect,2022
246,"Segatron-XL large, M=384 + HCP","Microsoft Research,University of Waterloo",3/21/2022,WikiText-103,Confident,Unreleased,Open (non-commercial),Operation counting,Indirect,2022
227,CoCa,Google Research,6/14/2022,"JFT-3B,ALIGN",Confident,Unreleased,Unreleased,Hardware,Indirect,2022
247,ViT-G (model soup),"University of Washington,Columbia University,Google,Meta AI,Tel Aviv University",3/10/2022,,Confident,Open weights (non-commercial),Unreleased,Operation counting,Indirect,2022
228,MetaLM,Microsoft Research,6/13/2022,The Pile,Unknown,,,,None,2022
229,DITTO,"Tsinghua University,Apple,Westlake University,Chinese University of Hong Kong (CUHK)",6/6/2022,WikiText-103,Confident,Unreleased,Open source,Operation counting,Indirect,2022
230,Diffusion-GAN,"UT Austin,Microsoft",6/5/2022,"CIFAR-10,LSUN Bedroom,AFHQ,LSUN Church,STL-10,FFHQ",Unknown,,,,None,2022
231,CogVideo,"Tsinghua University,Beijing Academy of Artificial Intelligence / BAAI",5/29/2022,Unspecified unreleased,Speculative,Open weights (unrestricted),Open source,Operation counting,Indirect,2022
233,Imagen,Google Brain,5/23/2022,"LAION-400M,Unspecified unreleased",Likely,API access,Unreleased,Hardware,None,2022
234,SimCSE,"Princeton University,Tsinghua University",5/18/2022,,Unknown,,,,None,2022
235,Gato,DeepMind,5/12/2022,,,Unreleased,Unreleased,"Hardware,Operation counting",None,2022
232,Tranception,"University of Oxford,Harvard Medical School,Cohere",5/27/2022,UniRef100,Confident,Open weights (unrestricted),,Hardware,Indirect,2022
237,DeBERTaV3large + KEAR,Microsoft,5/4/2022,,Confident,,,,Indirect,2022
238,OPT-175B,Meta AI,5/2/2022,"The Pile,BookCorpus (BooksCorpus, Toronto Book Corpus),CC-Stories,Pushshift Reddit",Confident,Open weights (non-commercial),Open source,Reported,Direct,2022
239,Flamingo,DeepMind,4/29/2022,"MultiModal MassiveWeb,LTIP,VTP,ALIGN",Confident,Unreleased,Unreleased,Hardware,Indirect,2022
240,Sparse all-MLP,Meta AI,4/14/2022,"RoBERTa dataset,CC100",,Unreleased,,Hardware,None,2022
241,Stable Diffusion (LDM-KL-8-G),"Runway,Ludwig Maximilian University",4/13/2022,LAION-400M,,Open weights (restricted use),,Hardware,Indirect,2022
242,BERT-RBP,Waseda University,4/7/2022,RBPSuite,Confident,Open weights (non-commercial),Open (non-commercial),Hardware,Indirect,2022
243,DALL·E 2,OpenAI,4/6/2022,"CLIP,DALL-E",Confident,,,,Indirect,2022
236,UL2,"Google Research,Google Brain",5/10/2022,C4,Confident,Open weights (unrestricted),,"Hardware,Operation counting",Indirect,2022
245,Chinchilla,DeepMind,3/29/2022,"MassiveWeb,C4",Confident,Unreleased,Unreleased,Reported,Indirect,2022
244,PaLM (540B),Google Research,4/4/2022,"Wikipedia,GLaM dataset,LaMBDA dataset,GitHub",Confident,Unreleased,Unreleased,Hardware,Direct,2022
104,RT-Trajectory,"Google DeepMind,University of California San Diego,Stanford University",11/3/2023,RT-1,Unknown,,,,None,2023
112,DALL·E 3,OpenAI,10/19/2023,Unspecified unreleased,Unknown,API access,Unreleased,,None,2023
110,DiT-XL/2 + CADS,ETH Zurich,10/26/2023,ImageNet,Likely,,,,None,2023
109,ChatGLM3-6B,Zhipu AI,10/27/2023,Unspecified unreleased,Likely,Open weights (restricted use),Unreleased,Operation counting,Indirect,2023
105,BLUUMI,"University of Turku,Hugging Face",11/3/2023,"Parsebank,mC4,Common Crawl,Wikipedia",Likely,Open weights (unrestricted),,,Indirect,2023
107,Cohere Embed,Cohere,11/2/2023,Unspecified unreleased,Unknown,API access,Unreleased,,None,2023
106,Yi-34B,01.AI,11/2/2023,Unspecified unreleased,Confident,Open weights (restricted use),Unreleased,Operation counting,Indirect,2023
113,ERNIE 4.0,Baidu,10/17/2023,,Unknown,,,,None,2023
108,Skywork-13B,Kunlun Inc.,10/30/2023,SkyPile,Confident,Open weights (restricted use),Open (restricted use),Operation counting,Indirect,2023
114,RT-2-X,Google DeepMind,10/13/2023,Open X-Embodiment,Confident,Unreleased,Unreleased,,Indirect,2023
124,Swift,Intel Labs,8/30/2023,,Likely,Unreleased,,Hardware,None,2023
116,FinGPT-13B,"University of California Los Angeles (UCLA),Columbia University,New York University (NYU)",10/7/2023,,Likely,Open weights (unrestricted),Open source,Hardware,Indirect,2023
117,CTM (CIFAR-10),"Stanford University,Sony",10/1/2023,CIFAR-10,Unknown,,,,None,2023
118,Amazon Titan,Amazon,9/28/2023,,Likely,API access,Unreleased,"Hardware,Operation counting",None,2023
119,Show-1,National University of Singapore,9/27/2023,WebVid-10M,Unknown,Open weights (non-commercial),Unreleased,,Indirect,2023
120,GPT-4V,OpenAI,9/25/2023,Unspecified unreleased,Unknown,API access,Unreleased,,None,2023
121,AlphaMissense,Google DeepMind,9/22/2023,"MGnify,UniRef90",Likely,Unreleased,Open source,,None,2023
122,Robot Parkour,"Shanghai Qi Zhi institute,Stanford University,Carnegie Mellon University (CMU),Tsinghua University",9/12/2023,,Confident,,,,Indirect,2023
123,Falcon-180B,Technology Innovation Institute,9/6/2023,RefinedWeb,Confident,Open weights (restricted use),Unreleased,"Reported,Operation counting",Indirect,2023
125,Jais,"Cerebras Systems,Mohamed bin Zayed University of Artificial Intelligence (MBZUAI),Inception",8/29/2023,"Abu El-Khair,Aranews,ArabicText 2022,C4 Arabic,Arabic Wikipedia,ArabicNews 2020,Maktabah,United Nations Parallel Corpus,The Pile,Books3,arXiv,PubMed Central,WebText2,English Wikipedia,FreeLaw,PubMed Abstracts,DeepMind Mathematics,Project Gutenberg,BookCorpus2,EuroParl,PhilPapers,YouTube Subtitles,NIH Grant Abstracts,Enron Emails,GitHub",Confident,Open weights (unrestricted),,Operation counting,Indirect,2023
126,PeptideBERT,Carnegie Mellon University (CMU),8/28/2023,,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2023
103,Grok-1,xAI,11/4/2023,Unspecified unreleased,Likely,Open weights (unrestricted),Unreleased,Benchmarks,Indirect,2023
115,Ferret (13B),"Columbia University,Apple",10/11/2023,GRIT,Confident,Open weights (non-commercial),Open (non-commercial),,Indirect,2023
102,LLaVA 1.5,"University of Wisconsin Madison,Microsoft Research",11/5/2023,Unspecified unreleased,Confident,Open weights (restricted use),,Hardware,Indirect,2023
81,Mixtral 8x7B,Mistral AI,12/11/2023,,Confident,Open weights (unrestricted),Unreleased,,Indirect,2023
100,GPT-4 Turbo,OpenAI,11/6/2023,Unspecified unreleased,Unknown,API access,Unreleased,Benchmarks,None,2023
76,CoRe,Tsinghua University,12/29/2023,"GSM8K,ASDiv",Speculative,,,,None,2023
77,Gemini Nano-2,Google DeepMind,12/19/2023,Unspecified unreleased,Confident,Unreleased,,,Indirect,2023
78,Gemini Nano-1,Google DeepMind,12/19/2023,Unspecified unreleased,Confident,Unreleased,,,Indirect,2023
79,FunSearch,Google DeepMind,12/14/2023,,Speculative,Open weights (unrestricted),Unreleased,Hardware,Indirect,2023
80,CogAgent,"Tsinghua University,Zhipu AI",12/14/2023,"COYO-700M,LAION-2B,Common Crawl,Unspecified unreleased",Likely,Open weights (restricted use),Open source,Operation counting,Indirect,2023
127,Qwen-VL,Alibaba,8/24/2023,,Likely,Open weights (restricted use),Unreleased,,Indirect,2023
82,SeamlessM4T,"Facebook,INRIA,University of California (UC) Berkeley",12/8/2023,,Confident,Open weights (unrestricted),Open source,,Indirect,2023
83,Llama Guard,Meta AI,12/7/2023,,Confident,Open weights (restricted use),Unreleased,Operation counting,Direct,2023
84,Gemini 1.0 Ultra,Google DeepMind,12/6/2023,Unspecified unreleased,Speculative,API access,Unreleased,"Benchmarks,Hardware",None,2023
85,Gemini 1.0 Pro,Google DeepMind,12/6/2023,Unspecified unreleased,Speculative,API access,Unreleased,Benchmarks,None,2023
86,Mamba-24M (SC09),"Carnegie Mellon University (CMU),Princeton University",12/1/2023,SC09,Confident,,,,Indirect,2023
101,CogVLM-17B,"Tsinghua University,Zhipu AI,Beihang University",11/6/2023,"VQAv2,LAION-2B,COYO-700M,OKVQA,TextVQA,OCR-VQA,ScienceQA,LLaVA-Instruct-150k,LRV-Instruction,LLaVAR,Flickr30K Entities,RefCOCO,Visual7W,VisualGenome,COCO,TextCaps",Confident,Open weights (restricted use),Unreleased,Reported,Indirect,2023
87,Qwen-72B,Alibaba,11/30/2023,,Confident,Open weights (restricted use),Unreleased,Operation counting,Indirect,2023
89,GNoME for crystal discovery,Google DeepMind,11/29/2023,,Likely,Unreleased,Unreleased,,None,2023
90,Inflection-2,Inflection AI,11/22/2023,Unspecified unreleased,Confident,Hosted access (no API),Unreleased,"Hardware,Benchmarks",Indirect,2023
91,Claude 2.1,Anthropic,11/21/2023,Unspecified unreleased,Unknown,API access,Unreleased,,None,2023
92,Nemotron-3-8B,NVIDIA,11/15/2023,"Unspecified unreleased,Flan,P3 (Public Pool of Prompts)",Confident,Open weights (restricted use),,"Operation counting,Hardware",Indirect,2023
93,Qwen-Audio-Chat,Alibaba,11/14/2023,,Likely,Open weights (restricted use),,,Indirect,2023
94,GraphCast,Google DeepMind,11/14/2023,,Speculative,Open weights (unrestricted),,Hardware,Indirect,2023
95,Volcano 13B,"Korea University,Korea Advanced Institute of Science and Technology (KAIST),LG",11/13/2023,"LAION,SBU,ShareGPT4V,Unspecified unreleased",Likely,Open weights (non-commercial),,Hardware,Indirect,2023
96,SPHINX (Llama 2 13B),"Shanghai AI Lab,Chinese University of Hong Kong (CUHK),ShanghaiTech University",11/13/2023,"LAION-400M,LAION-COCO,RefinedWeb",Likely,Open weights (restricted use),Open (restricted use),Hardware,None,2023
97,MultiBand Diffusion,"Meta AI,Hebrew University of Jerusalem,LORIA",11/8/2023,"Common Voice,DNS,MTG-Jamendo,FSD50K,AudioSet",Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2023
98,OmniVec,TensorTour,11/7/2023,"AudioSet,Something-Something v2 (SSv2),English Wikipedia,ImageNet-1k,SUN RGB-D,ModelNet40",Unknown,,,,None,2023
99,mPLUG-Owl2,Alibaba,11/7/2023,"Conceptual Captions (CC3M),Conceptual Captions 12M (CC12M),COCO,LAION,COYO-700M",Speculative,Open weights (unrestricted),,,Indirect,2023
88,PPLX-70B-Online,Perplexity,11/29/2023,,Likely,API access,,,None,2023
128,GGNN,"Westlake University,Tsinghua University,Toyota Technological Institute at Chicago",8/5/2023,,Confident,,,Other,Indirect,2023
111,CODEFUSION (Python),"Microsoft,Microsoft Research",10/26/2023,,Confident,,,Hardware,Indirect,2023
130,AudioLM,Google,7/26/2023,LibriLight,Speculative,,,Operation counting,None,2023
159,VideoMAE V2,"Nanjing University,Shenzhen Institute of Advanced Technology,Shanghai AI Lab",3/29/2023,,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2023
160,Firefly,Adobe,3/21/2023,Adobe Stock,Unknown,,,,None,2023
161,PanGu-Σ,Huawei Noah's Ark Lab,3/20/2023,,Confident,Unreleased,Unreleased,Hardware,Indirect,2023
162,Gen-2,Runway,3/20/2023,,Unknown,,,,None,2023
163,LEP-AD,"King Abdullah University of Science and Technology (KAUST),Karolinska Institute",3/15/2023,,Confident,Unreleased,Open (non-commercial),,Indirect,2023
164,GPT-4,OpenAI,3/15/2023,Unspecified unreleased,Speculative,API access,Unreleased,Hardware,None,2023
165,Falcon-40B,Technology Innovation Institute,3/15/2023,RefinedWeb,Confident,Open weights (unrestricted),Unreleased,"Operation counting,Reported",Indirect,2023
166,Claude,Anthropic,3/14/2023,Unspecified unreleased,Unknown,,,,None,2023
167,PaLM-E,"Google,TU Berlin",3/6/2023,,Likely,,,,Direct,2023
168,AudioGen,"Meta AI,Hebrew University of Jerusalem",3/5/2023,"AudioSet,AudioCaps",Likely,Open weights (non-commercial),Open source,Hardware,Indirect,2023
169,DiT-XL/2,"New York University (NYU),University of California (UC) Berkeley",3/2/2023,ImageNet,Confident,,,"Hardware,Other",Indirect,2023
170,LLaMA-65B,Meta AI,2/24/2023,"CCNet,GitHub,Wikipedia,books,arXiv,Stack Exchange",Confident,Open weights (non-commercial),Unreleased,Operation counting,Direct,2023
171,BASIC-L + Lion,"Google,University of California Los Angeles (UCLA)",2/13/2023,,Confident,,,,Indirect,2023
173,ProteinDT,"University of California (UC) Berkeley,California Institute of Technology,University of Toronto,University of Wisconsin Madison,Texas A&M,NVIDIA,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms)",2/9/2023,UniProtKB,Unknown,Unreleased,,,None,2023
174,Gen-1,Runway,2/6/2023,,Unknown,,,,None,2023
175,Flan T5-XXL + BLIP-2,Salesforce Research,1/30/2023,"COCO,LAION-400M",Confident,Open weights (unrestricted),Open source,,Direct,2023
176,BLIP-2 (Q-Former),Salesforce Research,1/30/2023,"COCO,LAION-400M,Conceptual Captions (CC3M),Conceptual Captions 12M (CC12M),VisualGenome,SBU",Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2023
177,DDPM-IP (CelebA),Utrecht University,1/27/2023,CelebA,Likely,,,Hardware,None,2023
178,MusicLM,Google,1/26/2023,Free Music Archive,Confident,,,,Indirect,2023
179,Ankh_large,"Technical University of Munich,Columbia University",1/16/2023,UniRef50,Confident,Open weights (non-commercial),,"Operation counting,Third-party estimation",Indirect,2023
180,Nucleotide Transformer,"NVIDIA,Technical University of Munich",1/15/2023,"Human Reference Genome (GRCh38/hg38),1000 Genomes Project",Likely,,,"Operation counting,Hardware",None,2023
181,VALL-E,Microsoft,1/5/2023,LibriLight,Speculative,Unreleased,,Operation counting,None,2023
129,RT-2,Google DeepMind,7/28/2023,RT-1,Confident,,,,Indirect,2023
158,BloombergGPT,"Bloomberg,Johns Hopkins University",3/30/2023,,Confident,Unreleased,Unreleased,"Reported,Hardware",None,2023
157,Segment Anything Model,Meta AI,4/5/2023,Segment Anything 1B,Confident,Open weights (unrestricted),Unreleased,Hardware,Indirect,2023
172,ViT-22B,Google,2/10/2023,JFT-4B,Confident,Unreleased,Unreleased,Hardware,Indirect,2023
155,DINOv2,"Facebook AI Research,INRIA",4/14/2023,,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2023
131,Llama 2-70B,Meta AI,7/18/2023,Llama 2 dataset,Confident,Open weights (restricted use),Unreleased,"Hardware,Operation counting",Direct,2023
156,Incoder-6.7B,"Facebook AI Research,University of Washington,University of California (UC) Berkeley,Carnegie Mellon University (CMU),Toyota Technological Institute at Chicago",4/9/2023,,Confident,Open weights (non-commercial),Unreleased,Reported,Indirect,2023
132,Llama 2-7B,Meta AI,7/18/2023,Llama 2 dataset,Confident,Open weights (restricted use),Unreleased,"Hardware,Operation counting",Direct,2023
133,Claude 2,Anthropic,7/11/2023,Unspecified unreleased,Speculative,API access,Unreleased,"Benchmarks,Hardware",None,2023
134,xTrimoPGLM -100B,"Tsinghua University,BioMap Research",7/6/2023,UniRef50,Confident,Unreleased,Unreleased,"Reported,Operation counting,Hardware",Indirect,2023
135,InternLM,"Shanghai AI Lab,SenseTime",7/6/2023,,Confident,,,Operation counting,Indirect,2023
137,Stable Diffusion XL (SDXL),Stability AI,7/4/2023,Unspecified unreleased,Speculative,,,,None,2023
138,HyenaDNA,"Stanford University,Harvard University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),University of Montreal / Université de Montréal",6/27/2023,Human Reference Genome (GRCh38/hg38),Confident,,,Hardware,Indirect,2023
139,ERNIE 3.5,Baidu,6/27/2023,,Unknown,,,,None,2023
140,RoboCat,"Google DeepMind,Google",6/20/2023,,Speculative,,,,None,2023
141,MusicGen,Meta AI,6/8/2023,ShutterStock and Pond5 music data collections,Likely,,,,None,2023
142,LTM-1,Magic,6/6/2023,,Unknown,,,,None,2023
136,Pangu-Weather,Huawei,7/5/2023,ERA5,Confident,Open weights (non-commercial),Unreleased,Hardware,Indirect,2023
144,Goat-7B,National University of Singapore,5/23/2023,,Speculative,Open weights (non-commercial),Open (non-commercial),,Indirect,2023
153,Agile Soccer Robot,Google DeepMind,4/26/2023,,Unknown,Unreleased,,,None,2023
143,PaLI-X,Google Research,5/29/2023,WebLI,Likely,,,,None,2023
152,ImageBind,Meta AI,5/9/2023,"SUN RGB-D,LLVIP,Ego4D,AudioSet",Likely,Open weights (non-commercial),Open (non-commercial),,Indirect,2023
151,StarCoder,"Hugging Face,ServiceNow,Northeastern University,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms),Carnegie Mellon University (CMU),Johns Hopkins University,Leipzig University,ScaDS.AI,Queen Mary University of London,Roblox,Sea AI Lab,Technion - Israel Institute of Technology,Monash University,CSIRO,Data61,McGill University,Saama,University of British Columbia (UBC),Massachusetts Institute of Technology (MIT),Technical University of Munich,IBM,University of Vermont,UnfoldML,SAP,University of Notre Dame,Columbia University,New York University (NYU),University of Allahabad,Discover Dollar,Toloka,Telefonica,Stanford University,Weizmann Institute of Science,Alan Turing Institute,Wellesley College,EleutherAI,Forschungszentrum Julich",5/9/2023,The Stack,Confident,Open weights (restricted use),Unreleased,"Reported,Hardware",Indirect,2023
149,InstructBLIP,"Salesforce Research,Hong Kong University of Science and Technology,Nanyang Technological University",5/11/2023,"COCO,Web CapFilt,NoCaps,Flickr30K Entities,TextCaps,VQAv2,VizWiz,GQA,OKVQA,ScienceQA,OCR-VQA,TextVQA,LLaVA-Instruct-150k",Confident,Open weights (non-commercial),,Hardware,Indirect,2023
150,PaLM 2,Google,5/10/2023,,Likely,API access,Unreleased,Operation counting,Indirect,2023
148,Med-PaLM 2,"Google Research,DeepMind",5/16/2023,MultiMedQA,Likely,Unreleased,Unreleased,,Indirect,2023
147,CoEdiT-xxl,"University of Minnesota,Grammarly",5/17/2023,,Likely,Open weights (non-commercial),Open (non-commercial),,Indirect,2023
146,ONE-PEACE,"Alibaba,Huazhong University of Science and Technology",5/18/2023,"LAION-2B,LAION-Audio-630K",Speculative,Open weights (unrestricted),Open source,Operation counting,Indirect,2023
145,CodeT5+,Salesforce,5/20/2023,,,Open weights (unrestricted),,,Direct,2023
154,LLaVA,"University of Wisconsin Madison,Microsoft Research,Columbia University",4/17/2023,Conceptual Captions (CC3M),Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2023
31,Qwen2.5 Instruct (72B),Alibaba,9/19/2024,Unspecified unreleased,Confident,Open weights (restricted use),,Operation counting,Indirect,2024
28,Palmyra X 004,Writer,10/9/2024,,,API access,,,None,2024
29,Movie Gen Video,Meta AI,10/4/2024,,Confident,Unreleased,,Operation counting,Indirect,2024
27,CHAI-1,Chai discovery,10/15/2024,"PDB (Protein Data Bank), AlphaFold database (AFDB)",Confident,Open weights (non-commercial),Open (non-commercial),Hardware,Indirect,2024
30,Qwen2.5-72B,Alibaba,9/19/2024,Unspecified unreleased,Confident,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2024
32,Qwen2.5-32B,Alibaba,9/17/2024,Unspecified unreleased,Confident,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2024
36,Hunyuan Turbo,Tencent,9/5/2024,Unspecified unreleased,Unknown,,,,None,2024
34,o1-mini,OpenAI,9/12/2024,Unspecified unreleased,Unknown,API access,Unreleased,,None,2024
35,DeepSeek-V2.5,DeepSeek,9/6/2024,"GitHub,Common Crawl",Confident,Open weights (restricted use),Unreleased,Operation counting,Indirect,2024
37,AlphaProteo,Google DeepMind,9/5/2024,PDB (Protein Data Bank),Unknown,Unreleased,Unreleased,,None,2024
38,GLM-4-Plus,Zhipu AI,8/29/2024,,Unknown,API access,,Benchmarks,None,2024
26,Yi-Lightning,01.AI,10/18/2024,Unspecified unreleased,Confident,API access,Unreleased,Hardware,Indirect,2024
39,Jamba 1.5-Large,AI21 Labs,8/22/2024,Unspecified unreleased,Confident,Open weights (restricted use),Unreleased,,Indirect,2024
33,o1-preview,OpenAI,9/12/2024,Unspecified unreleased,Unknown,API access,Unreleased,,None,2024
25,NVLM-D 72B,NVIDIA,10/22/2024,"COCO,Conceptual Captions (CC3M),SBU,VQAv2,VisualGenome,TextVQA,OCR-VQA",Confident,Open weights (non-commercial),Open (non-commercial),Operation counting,Indirect,2024
13,Gemini 2.0 Pro,"Google DeepMind,Google",12/11/2024,Unspecified unreleased,Unknown,Hosted access (no API),Unreleased,,None,2024
23,NVLM-X 72B,NVIDIA,10/22/2024,"COCO,Conceptual Captions (CC3M),SBU,VQAv2,VisualGenome,TextVQA,OCR-VQA",Likely,Open weights (non-commercial),,Operation counting,Indirect,2024
22,Doubao-pro,ByteDance,10/28/2024,Unspecified unreleased,Speculative,API access,Unreleased,Operation counting,None,2024
21,Hunyuan-Large,Tencent,11/6/2024,Unspecified unreleased,Confident,Open weights (restricted use),Open (restricted use),Operation counting,Indirect,2024
20,Pixtral Large,Mistral AI,11/18/2024,,Confident,Open weights (restricted use),,,Indirect,2024
19,Suno v4,Suno,11/19/2024,,Unknown,API access,,,None,2024
18,Fugatto 1,NVIDIA,11/25/2024,,Confident,Unreleased,,,Indirect,2024
17,Amazon Nova Pro,Amazon,12/3/2024,,Speculative,API access,,Comparison with other models,None,2024
16,o1,OpenAI,12/5/2024,Unspecified unreleased,Unknown,API access,Unreleased,,None,2024
15,Llama 3.3,Meta AI,12/6/2024,Unspecified unreleased,Confident,Open weights (restricted use),Unreleased,"Operation counting,Hardware",Direct,2024
14,EXAONE 3.5 32B,LG AI Research,12/9/2024,Unspecified unreleased,Confident,Open weights (non-commercial),Unreleased,Reported,Indirect,2024
12,Veo 2,Google DeepMind,12/16/2024,Unspecified unreleased,Unknown,API access,,,None,2024
11,o3,OpenAI,12/20/2024,Unspecified unreleased,Unknown,Unreleased,Unreleased,,None,2024
10,DeepSeek-V3,DeepSeek,12/24/2024,,Confident,Open weights (restricted use),,"Operation counting,Hardware",Indirect,2024
40,Grok-2,xAI,8/13/2024,Unspecified unreleased,Confident,Hosted access (no API),Unreleased,"Comparison with other models,Reported",Indirect,2024
24,NVLM-H 72B,NVIDIA,10/22/2024,"COCO,Conceptual Captions (CC3M),SBU,VQAv2,VisualGenome,TextVQA,OCR-VQA",Likely,Open weights (non-commercial),,Operation counting,Indirect,2024
41,Table Tennis Agent,Google DeepMind,8/7/2024,,Likely,Unreleased,Unreleased,,None,2024
63,Claude 3 Opus,Anthropic,3/4/2024,Unspecified unreleased,Speculative,API access,Unreleased,Benchmarks,None,2024
43,AFM-on-device,Apple,7/29/2024,,Confident,Hosted access (no API),Unreleased,Operation counting,Indirect,2024
75,Kimi Explorer,Moonshot,1/1/2024,,Unknown,,,,None,2024
74,Palmyra X 003,Writer,1/1/2024,,,API access,,,None,2024
73,AlphaGeometry,"Google DeepMind,New York University (NYU)",1/17/2024,,Confident,Open weights (unrestricted),Open source,,Indirect,2024
72,Qwen-VL-Max,Alibaba,1/25/2024,Unspecified unreleased,Confident,API access,,,Indirect,2024
71,Qwen1.5-72B,Alibaba,2/4/2024,Unspecified unreleased,Confident,Open weights (restricted use),Unreleased,Operation counting,Indirect,2024
70,Aya,"Cohere for AI,Brown University,Cohere,Carnegie Mellon University (CMU),Massachusetts Institute of Technology (MIT)",2/12/2024,,Speculative,Open weights (unrestricted),Unreleased,,Indirect,2024
69,Gemini 1.5 Pro,Google DeepMind,2/15/2024,Unspecified unreleased,Speculative,API access,Unreleased,Benchmarks,None,2024
68,Sora,OpenAI,2/15/2024,Unspecified unreleased,Unknown,Unreleased,Unreleased,,None,2024
67,Sora Turbo,OpenAI,2/15/2024,Unspecified unreleased,Unknown,Unreleased,Unreleased,,None,2024
66,MegaScale (Production),"ByteDance,Peking University",2/23/2024,,Speculative,Unreleased,Unreleased,Other,None,2024
42,AFM-server,Apple,7/29/2024,,Likely,Hosted access (no API),Unreleased,"Operation counting,Hardware",None,2024
64,Aramco Metabrain AI,Saudi Aramco,3/4/2024,,Likely,Unreleased,,Operation counting,None,2024
62,Claude 3 Sonnet,Anthropic,3/4/2024,Unspecified unreleased,Unknown,API access,Unreleased,,None,2024
61,Inflection-2.5,Inflection AI,3/7/2024,,Speculative,Hosted access (no API),Unreleased,Comparison with other models,None,2024
60,MM1-30B,Apple,3/14/2024,"Conceptual Captions (CC3M),Conceptual Captions 12M (CC12M),COYO-700M,Unspecified unreleased,OBELICS",Likely,Unreleased,Unreleased,Operation counting,None,2024
65,Mistral Large,Mistral AI,2/26/2024,,Likely,API access,Unreleased,Cost,None,2024
58,ReALM,Apple,3/29/2024,,Confident,Unreleased,,,Indirect,2024
44,Mistral Large 2,Mistral AI,7/24/2024,Unspecified unreleased,Likely,Open weights (non-commercial),Unreleased,"Hardware,Cost,Benchmarks",Indirect,2024
45,Llama 3.1-405B,Meta AI,7/23/2024,Llama 3 dataset,Confident,Open weights (restricted use),Open (restricted use),"Reported,Operation counting",Direct,2024
59,DBRX,Databricks,3/27/2024,,Confident,Open weights (restricted use),Unreleased,Operation counting,Indirect,2024
47,ESM3 (98B),"EvolutionaryScale,University of California (UC) Berkeley",6/25/2024,ESM3 Dataset,Confident,Unreleased,Unreleased,Reported,Indirect,2024
48,Claude 3.5 Sonnet,Anthropic,6/20/2024,Unspecified unreleased,Speculative,API access,Unreleased,Benchmarks,None,2024
49,DeepSeek-Coder-V2 236B,DeepSeek,6/17/2024,"GitHub,Common Crawl",Confident,Open weights (restricted use),Unreleased,Operation counting,Indirect,2024
50,Nemotron-4 340B,NVIDIA,6/14/2024,Unspecified unreleased,Confident,Open weights (unrestricted),Unreleased,"Operation counting,Hardware",Indirect,2024
46,GPT-4o mini,OpenAI,7/18/2024,Unspecified unreleased,Speculative,API access,Unreleased,Benchmarks,None,2024
52,Qwen2-72B,Alibaba,6/7/2024,Unspecified unreleased,Confident,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2024
53,GLM-4 (0520),Zhipu AI,5/20/2024,,Likely,API access,,Operation counting,None,2024
54,Yi-Large,01.AI,5/13/2024,,Speculative,API access,Unreleased,Operation counting,None,2024
55,GPT-4o,OpenAI,5/13/2024,Unspecified unreleased,Speculative,API access,Unreleased,Benchmarks,None,2024
56,Llama 3-70B,Meta AI,4/18/2024,Llama 3 dataset,Confident,Open weights (restricted use),Unreleased,"Operation counting,Hardware",Direct,2024
57,Reka Core,Reka AI,4/15/2024,"Wikipedia,Unspecified unreleased",Speculative,API access,Unreleased,Hardware,None,2024
51,OpenVLA,"Stanford University,University of California (UC) Berkeley,Toyota Research Institute,Google DeepMind,Massachusetts Institute of Technology (MIT),Physical Intelligence",6/13/2024,Open X-Embodiment,Confident,Open weights (unrestricted),Open source,Hardware,Indirect,2024
1,QwQ-32B,Alibaba,3/6/2025,Unspecified unreleased,Speculative,Open weights (unrestricted),Unreleased,,Indirect,2025
2,GPT-4.5,OpenAI,2/27/2025,Unspecified unreleased,Unknown,API access,Unreleased,,None,2025
3,Claude 3.7 Sonnet,Anthropic,2/24/2025,Unspecified unreleased,Likely,API access,Unreleased,,None,2025
4,Grok-3,xAI,2/17/2025,Unspecified unreleased,Confident,Hosted access (no API),Unreleased,"Hardware,Comparison with other models",Indirect,2025
7,Kimi k1.5,Moonshot,1/22/2025,Unspecified unreleased,Unknown,API access,Unreleased,,None,2025
6,Computer-Using Agent (CUA),OpenAI,1/23/2025,Unspecified unreleased,Unknown,Hosted access (no API),Unreleased,,None,2025
8,Doubao-1.5-pro,ByteDance,1/22/2025,,Unknown,Hosted access (no API),Unreleased,,None,2025
9,DeepSeek-R1,DeepSeek,1/20/2025,Unspecified unreleased,Confident,Open weights (unrestricted),Unreleased,Operation counting,Indirect,2025
5,o3-mini,OpenAI,1/31/2025,Unspecified unreleased,Unknown,API access,Unreleased,,None,2025
0,EXAONE Deep 32B,LG AI Research,3/16/2025,Unspecified unreleased,Confident,Open weights (non-commercial),Unreleased,"Reported,Operation counting,Hardware",Indirect,2025