{
"models": [
{
"name": "ppocrv5_det",
"model_dir": "../inference_models/PP-OCRv5_server_det_infer",
"default_opsets": [17],
"input_kind": "det",
"sample_inputs": {
"x": {
"shape": [1, 3, 640, 640],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "ppocrv5_rec",
"model_dir": "../inference_models/PP-OCRv5_server_rec_infer",
"default_opsets": [17],
"input_kind": "rec",
"sample_inputs": {
"x": {
"shape": [1, 3, 48, 320],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "slanet",
"model_dir": "../inference_models/SLANet_infer",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 3, 488, 488],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "ppdoclayout_plus_l",
"model_dir": "../inference_models/PP-DocLayout_plus-L_infer",
"default_opsets": [17],
"input_kind": "layout",
"diff_tolerance": {
"rtol": 0.001,
"atol": 0.005
},
"sample_inputs": {
"image": {
"shape": [1, 3, 800, 800],
"dtype": "float32",
"mode": "randn"
},
"im_shape": {
"value": [[800.0, 800.0]],
"dtype": "float32"
},
"scale_factor": {
"value": [[1.0, 1.0]],
"dtype": "float32"
}
}
},
{
"name": "ppdoclayoutv2",
"model_dir": "../inference_models/PP-DocLayoutV2_infer",
"semantic_baseline": "baselines/ppdoclayoutv2_opset17_seed20260412_semantic.npz",
"default_opsets": [17],
"input_kind": "layout",
"diff_tolerance": {
"rtol": 0.001,
"atol": 0.005
},
"sample_inputs": {
"image": {
"shape": [1, 3, 800, 800],
"dtype": "float32",
"mode": "randn"
},
"im_shape": {
"value": [[800.0, 800.0]],
"dtype": "float32"
},
"scale_factor": {
"value": [[1.0, 1.0]],
"dtype": "float32"
}
},
"compare_config": {
"mode": "instance_row_match",
"key_output": 0,
"reorder_outputs": [0],
"sort_columns": [0, 1, 2, 3, 4, 5],
"match_columns": [0, 1, 2, 3, 4, 5],
"exact_columns": [0],
"round_decimals": 6,
"per_output": {
"0": {
"permutation_columns": [6, 7]
}
}
}
},
{
"name": "ppdoclayoutv3",
"model_dir": "../inference_models/PP-DocLayoutV3_infer",
"default_opsets": [17],
"input_kind": "layout",
"diff_tolerance": {
"rtol": 0.001,
"atol": 0.005
},
"sample_inputs": {
"image": {
"shape": [1, 3, 800, 800],
"dtype": "float32",
"mode": "randn"
},
"im_shape": {
"value": [[800.0, 800.0]],
"dtype": "float32"
},
"scale_factor": {
"value": [[1.0, 1.0]],
"dtype": "float32"
}
},
"compare_config": {
"mode": "instance_row_match",
"key_output": 0,
"reorder_outputs": [0, 2],
"sort_columns": [0, 1, 2, 3, 4, 5],
"match_columns": [0, 1, 2, 3, 4, 5],
"exact_columns": [0],
"tie_group_columns": [0, 1, 2, 3, 4, 5],
"tie_aware_outputs": [2],
"round_decimals": 6,
"per_output": {
"0": {
"permutation_columns": [6]
},
"2": {
"int_tolerance": {
"max_abs": 1,
"mean_abs": 1e-5
}
}
}
}
},
{
"name": "slanext_wired",
"model_dir": "../inference_models/SLANeXt_wired_infer",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 3, 512, 512],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "slanext_wireless",
"model_dir": "../inference_models/SLANeXt_wireless_infer",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 3, 512, 512],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "ppformulanet_l",
"model_dir": "../inference_models/PP-FormulaNet-L_infer",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 1, 768, 768],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "ppformulanet_plus_l",
"model_dir": "../inference_models/PP-FormulaNet_plus-L_infer",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 1, 768, 768],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "latex_ocr",
"model_dir": "../inference_models/LaTeX_OCR_rec_infer",
"semantic_baseline": "baselines/latex_ocr_opset17_seed20260412_semantic.npz",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 1, 192, 672],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "uvdoc",
"model_dir": "../inference_models/UVDoc_infer",
"default_opsets": [17],
"diff_tolerance": {
"rtol": 0.001,
"atol": 0.005
},
"sample_inputs": {
"image": {
"shape": [1, 3, 256, 128],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "unimernet",
"model_dir": "../inference_models/UniMERNet_infer",
"default_opsets": [17],
"sample_inputs": {
"x": {
"shape": [1, 1, 192, 672],
"dtype": "float32",
"mode": "randn"
}
}
},
{
"name": "picodet_layout_1x",
"model_dir": "../inference_models/PicoDet_layout_1x_infer",
"default_opsets": [17],
"input_kind": "layout",
"sample_inputs": {
"image": {
"shape": [1, 3, 800, 608],
"dtype": "float32",
"mode": "randn"
},
"scale_factor": {
"value": [[1.0, 1.0]],
"dtype": "float32"
}
}
},
{
"name": "rtdetr_h_layout_17cls",
"model_dir": "../inference_models/RT-DETR-H_layout_17cls_infer",
"default_opsets": [17],
"input_kind": "layout",
"diff_tolerance": {
"rtol": 0.001,
"atol": 0.005
},
"sample_inputs": {
"image": {
"shape": [1, 3, 640, 640],
"dtype": "float32",
"mode": "randn"
},
"im_shape": {
"value": [[640.0, 640.0]],
"dtype": "float32"
},
"scale_factor": {
"value": [[1.0, 1.0]],
"dtype": "float32"
}
},
"compare_config": {
"mode": "instance_row_match",
"key_output": 0,
"reorder_outputs": [0],
"sort_columns": [0, 2, 3, 4, 5, 1],
"match_columns": [0, 2, 3, 4, 5],
"exact_columns": [0],
"round_decimals": 4
}
},
{
"name": "rtdetr_l_wired_table_cell_det",
"model_dir": "../inference_models/RT-DETR-L_wired_table_cell_det_infer",
"default_opsets": [17],
"input_kind": "layout",
"diff_tolerance": {
"rtol": 0.001,
"atol": 0.005
},
"sample_inputs": {
"image": {
"shape": [1, 3, 640, 640],
"dtype": "float32",
"mode": "randn"
},
"im_shape": {
"value": [[640.0, 640.0]],
"dtype": "float32"
},
"scale_factor": {
"value": [[1.0, 1.0]],
"dtype": "float32"
}
},
"compare_config": {
"mode": "instance_row_match",
"key_output": 0,
"reorder_outputs": [0],
"sort_columns": [0, 2, 3, 4, 5, 1],
"match_columns": [0, 2, 3, 4, 5],
"exact_columns": [0],
"round_decimals": 4
}
}
]
}