AMD-SHARK-Studio/tank/torch_model_list.csv

1	model_name	use_tracing	model_type	dynamic	mlir_type	decompose	param_count	tags	notes
2	efficientnet_b0	True	vision	False	linalg	False	5.3M	image-classification;cnn;conv2d;depthwise-conv	Smallest EfficientNet variant with 224x224 input
3	efficientnet_b7	True	vision	False	linalg	False	66M	image-classification;cnn;conv2d;depthwise-conv	Largest EfficientNet variant with 600x600 input
4	microsoft/MiniLM-L12-H384-uncased	True	hf	True	linalg	False	66M	nlp;bert-variant;transformer-encoder	Large version has 12 layers; 384 hidden size; Smaller than BERTbase (66M params vs 109M params)
5	bert-base-uncased	True	hf	True	linalg	False	109M	nlp;bert-variant;transformer-encoder	12 layers; 768 hidden; 12 attention heads
6	bert-base-cased	True	hf	True	linalg	False	109M	nlp;bert-variant;transformer-encoder	12 layers; 768 hidden; 12 attention heads
7	google/mobilebert-uncased	True	hf	True	linalg	False	25M	nlp,bert-variant,transformer-encoder,mobile	24 layers, 512 hidden size, 128 embedding
8	alexnet	False	vision	True	linalg	False	61M	cnn,parallel-layers	The CNN that revolutionized computer vision (move away from hand-crafted features to neural networks),10 years old now and probably no longer used in prod.
9	resnet18	False	vision	True	linalg	False	11M	cnn,image-classification,residuals,resnet-variant	1 7x7 conv2d and the rest are 3x3 conv2d
10	resnet50	False	vision	True	linalg	False	23M	cnn,image-classification,residuals,resnet-variant	Bottlenecks with only conv2d (1x1 conv -> 3x3 conv -> 1x1 conv blocks)
11	resnet101	False	vision	True	linalg	False	29M	cnn,image-classification,residuals,resnet-variant	Bottlenecks with only conv2d (1x1 conv -> 3x3 conv -> 1x1 conv blocks)
12	squeezenet1_0	False	vision	True	linalg	False	1.25M	cnn,image-classification,mobile,parallel-layers	Parallel conv2d (1x1 conv to compress -> (3x3 expand | 1x1 expand) -> concat)
13	wide_resnet50_2	False	vision	True	linalg	False	69M	cnn,image-classification,residuals,resnet-variant	Resnet variant where model depth is decreased and width is increased.
14	mobilenet_v3_small	False	vision	True	linalg	False	2.5M	image-classification,cnn,mobile	N/A
15	google/vit-base-patch16-224	True	hf_img_cls	False	linalg	False	86M	image-classification,vision-transformer,transformer-encoder	N/A
16	microsoft/resnet-50	True	hf_img_cls	False	linalg	False	23M	image-classification,cnn,residuals,resnet-variant	Bottlenecks with only conv2d (1x1 conv -> 3x3 conv -> 1x1 conv blocks)
17	facebook/deit-small-distilled-patch16-224	True	hf_img_cls	False	linalg	False	22M	image-classification,vision-transformer,cnn	N/A
18	microsoft/beit-base-patch16-224-pt22k-ft22k	True	hf_img_cls	False	linalg	False	86M	image-classification,transformer-encoder,bert-variant,vision-transformer	N/A
19	nvidia/mit-b0	True	hf_img_cls	False	linalg	False	3.7M	image-classification,transformer-encoder	SegFormer
20	mnasnet1_0	False	vision	True	linalg	False	-	cnn, torchvision, mobile, architecture-search	Outperforms other mobile CNNs on Accuracy vs. Latency
21	resnet50_fp16	False	vision	True	linalg	False	23M	cnn,image-classification,residuals,resnet-variant	Bottlenecks with only conv2d (1x1 conv -> 3x3 conv -> 1x1 conv blocks)
22	bert-base-uncased_fp16	True	fp16	False	linalg	False	109M	nlp;bert-variant;transformer-encoder	12 layers; 768 hidden; 12 attention heads
23	bert-large-uncased	True	hf	True	linalg	False	330M	nlp;bert-variant;transformer-encoder	24 layers, 1024 hidden units, 16 attention heads
24	bert-base-uncased	True	hf	False	stablehlo	False	109M	nlp;bert-variant;transformer-encoder	12 layers; 768 hidden; 12 attention heads
25	gpt2	True	hf_causallm	False	stablehlo	True	125M	nlp;transformer-encoder	-
26	facebook/opt-125m	True	hf	False	stablehlo	True	125M	nlp;transformer-encoder	-
27	distilgpt2	True	hf	False	stablehlo	True	88M	nlp;transformer-encoder	-
28	microsoft/deberta-v3-base	True	hf	False	stablehlo	True	88M	nlp;transformer-encoder	-