{ "format": "LLM-SCAN", "format_version": "1.0", "generator": { "name": "llmscan-engine", "version": "2.0.0" }, "scan_context": { "started_at": "2026-07-09T10:30:38+00:00Z", "ended_at": "2026-07-09T10:30:38+00:00Z", "duration_ms": 10925, "limits": { "timeout_ms": 60000, "sandbox": true }, "mode": "static-deep" }, "source": { "input_type": "file", "filename": "sense-voice-encoder.onnx", "size_bytes": 937179517, "hashes": { "sha256": "b1145eb7e9d2dbb9005fc99ed0313efebfadffea4aee1f4fa68b69a359126227" } }, "model": { "format": "onnx", "arch": null, "quantization": null, "params_estimate": 233990207, "file_size_bytes": 937179517, "sha256": "b1145eb7e9d2dbb9005fc99ed0313efebfadffea4aee1f4fa68b69a359126227", "sha512": "277a42c8b1dc47f55df706d9b5bf51b43324b792c628cdb149103d9145cd773173b0eb1eecb34952d389a999de2b4d57dc6703b305480b43d8f439885ba0ab70", "kv_meta": {}, "scanner_model_info": { "producer_name": "pytorch", "producer_version": "2.3.0", "domain": "", "model_version": 0, "doc_string": null, "ir_version": 8, "opsets": { "ai.onnx": 17 }, "node_count": 7496, "unique_op_types": 28, "op_histogram": { "Constant": 2405, "Add": 632, "Unsqueeze": 566, "Shape": 563, "Gather": 563, "Transpose": 490, "Reshape": 423, "MatMul": 421, "Concat": 352, "Cast": 222, "Mul": 214, "LayerNormalization": 142, "Split": 70, "ConstantOfShape": 70, "Slice": 70, "Pad": 70, "Conv": 70, "Softmax": 70, "Relu": 70, "Range": 3, "Sub": 2, "Div": 2, "ReduceMax": 1, "Less": 1, "Neg": 1, "Exp": 1, "Sin": 1, "Cos": 1 }, "op_risk_report": { "rows": [ { "op": "ConstantOfShape", "count": 70, "risk": "memory_amp", "note": "Memory amplification — allocates tensor of declared shape" }, { "op": "Gather", "count": 563, "risk": "runtime_sensitive", "note": "Runtime-sensitive — index-based gather" }, { "op": "Constant", "count": 2405, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Add", "count": 632, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Unsqueeze", "count": 566, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Shape", "count": 563, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Transpose", "count": 490, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Reshape", "count": 423, "risk": "normal", "note": "Standard ONNX op" }, { "op": "MatMul", "count": 421, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Concat", "count": 352, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Cast", "count": 222, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Mul", "count": 214, "risk": "normal", "note": "Standard ONNX op" }, { "op": "LayerNormalization", "count": 142, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Split", "count": 70, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Slice", "count": 70, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Pad", "count": 70, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Conv", "count": 70, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Softmax", "count": 70, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Relu", "count": 70, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Range", "count": 3, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Sub", "count": 2, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Div", "count": 2, "risk": "normal", "note": "Standard ONNX op" }, { "op": "ReduceMax", "count": 1, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Less", "count": 1, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Neg", "count": 1, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Exp", "count": 1, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Sin", "count": 1, "risk": "normal", "note": "Standard ONNX op" }, { "op": "Cos", "count": 1, "risk": "normal", "note": "Standard ONNX op" } ], "elevated_categories": [ "memory_amp", "runtime_sensitive" ], "has_control_flow": false, "has_memory_amp": true, "has_provider_compat": false, "has_runtime_sensitive": true, "has_quantized": false, "has_nondeterministic": false, "has_string_ops": false }, "inputs": [ { "name": "speech", "dtype": "FLOAT", "shape": [ 1, "speech_length", 560 ] }, { "name": "speech_lengths", "dtype": "INT64", "shape": [ 1 ] } ], "outputs": [ { "name": "encoder_out", "dtype": "FLOAT", "shape": [ "Addencoder_out_dim_0", "encoder_out_length", 25055 ] } ], "shape_risk_report": { "inputs": [ { "name": "speech", "dtype": "FLOAT", "shape": [ 1, "speech_length", 560 ], "has_dynamic": true, "has_negative": false, "est_elements": 560 }, { "name": "speech_lengths", "dtype": "INT64", "shape": [ 1 ], "has_dynamic": false, "has_negative": false, "est_elements": 1 } ], "outputs": [ { "name": "encoder_out", "dtype": "FLOAT", "shape": [ "Addencoder_out_dim_0", "encoder_out_length", 25055 ], "has_dynamic": true, "has_negative": false, "est_elements": 25055 } ], "any_dynamic_dims": true, "dynamic_dim_names": [ "Addencoder_out_dim_0", "encoder_out_length", "speech_length" ], "negative_dims_detected": false, "extreme_dims_detected": false, "input_elements_est": 561, "output_elements_est": 25055, "amplification_ratio": 44.66 }, "tensor_count": 916, "param_estimate": 233990207, "weights_size_bytes": 935960828, "weights_size_mb": 892.6, "dtype_histogram": { "FLOAT": 233990207 }, "custom_domains": [], "custom_domain_detail": {}, "custom_ops_report": { "has_non_standard": false, "standard_only": true, "domains": [] }, "has_external_data": false, "external_refs": [], "external_data_checks": { "present": false, "tensor_count": 0, "absolute_paths": [], "traversal_detected": [], "windows_drive_paths": [], "remote_url_paths": [], "null_byte_paths": [], "duplicate_files": [], "missing_files": [], "oob_reads": [], "size_validated_count": 0, "size_validation_total": 0 }, "quantized": false, "model_type": "audio-speech", "metadata_props": {}, "inline_tensor_mismatches": 0, "zero_dim_initializers": 0, "extreme_dim_tensors": 0, "initializer_integrity": { "count": 916, "inline_tensor_mismatches": 0, "empty_initializers": 0, "duplicate_names": 0, "duplicate_name_list": [], "unused_by_graph": 0, "unused_names": [], "shadowing_graph_inputs": 0, "shadowing_names": [], "nan_inf_scan": { "performed": true, "method": "sampled", "tensors_scanned": 916, "total_float_tensors": 916, "values_scanned": 14937359, "nan_count": 0, "inf_count": 0, "affected_tensors": [] } }, "declared_weight_bytes": 935960828, "declared_weight_mb": 892.6, "weight_delta_bytes": 1218689, "weight_delta_mb": 1.16, "file_size_bytes": 937179517, "file_size_mb": 893.76, "sha256": "b1145eb7e9d2dbb9005fc99ed0313efebfadffea4aee1f4fa68b69a359126227", "sha512": "277a42c8b1dc47f55df706d9b5bf51b43324b792c628cdb149103d9145cd773173b0eb1eecb34952d389a999de2b4d57dc6703b305480b43d8f439885ba0ab70", "dtype_stats": { "FLOAT": { "count": 916, "bytes": 935960828, "mb": 892.6, "avg_bits_per_elem": 32.0 } }, "tensor_preview": [ { "index": 0, "name": "model.encoders0.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 1, "name": "model.encoders0.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 2, "name": "model.encoders0.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 3, "name": "model.encoders0.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 4, "name": "model.encoders0.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 5, "name": "model.encoders.0.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 6, "name": "model.encoders.0.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 7, "name": "model.encoders.0.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 8, "name": "model.encoders.0.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 9, "name": "model.encoders.0.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 10, "name": "model.encoders.1.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 11, "name": "model.encoders.1.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 12, "name": "model.encoders.1.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 13, "name": "model.encoders.1.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 14, "name": "model.encoders.1.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 15, "name": "model.encoders.2.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 16, "name": "model.encoders.2.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 17, "name": "model.encoders.2.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 18, "name": "model.encoders.2.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 19, "name": "model.encoders.2.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 20, "name": "model.encoders.3.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 21, "name": "model.encoders.3.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 22, "name": "model.encoders.3.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 23, "name": "model.encoders.3.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 24, "name": "model.encoders.3.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 25, "name": "model.encoders.4.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 26, "name": "model.encoders.4.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 27, "name": "model.encoders.4.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 28, "name": "model.encoders.4.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 29, "name": "model.encoders.4.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 30, "name": "model.encoders.5.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 31, "name": "model.encoders.5.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 32, "name": "model.encoders.5.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 33, "name": "model.encoders.5.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 34, "name": "model.encoders.5.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 35, "name": "model.encoders.6.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 36, "name": "model.encoders.6.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 37, "name": "model.encoders.6.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 38, "name": "model.encoders.6.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 39, "name": "model.encoders.6.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 40, "name": "model.encoders.7.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 41, "name": "model.encoders.7.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 42, "name": "model.encoders.7.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 43, "name": "model.encoders.7.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 44, "name": "model.encoders.7.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 45, "name": "model.encoders.8.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 46, "name": "model.encoders.8.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 47, "name": "model.encoders.8.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 48, "name": "model.encoders.8.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 49, "name": "model.encoders.8.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 } ], "op_histogram_full": { "Constant": 2405, "Add": 632, "Unsqueeze": 566, "Shape": 563, "Gather": 563, "Transpose": 490, "Reshape": 423, "MatMul": 421, "Concat": 352, "Cast": 222, "Mul": 214, "LayerNormalization": 142, "Split": 70, "ConstantOfShape": 70, "Slice": 70, "Pad": 70, "Conv": 70, "Softmax": 70, "Relu": 70, "Range": 3, "Sub": 2, "Div": 2, "ReduceMax": 1, "Less": 1, "Neg": 1, "Exp": 1, "Sin": 1, "Cos": 1 } }, "opsets": { "ai.onnx": 17 }, "producer": "pytorch", "domain": "", "ir_version": 8, "nodes": 7496, "initialized_tensors": 916, "weights_size_est_bytes": 935960828, "inputs": [ { "name": "speech", "dtype": "FLOAT", "shape": [ 1, "speech_length", 560 ] }, { "name": "speech_lengths", "dtype": "INT64", "shape": [ 1 ] } ], "outputs": [ { "name": "encoder_out", "dtype": "FLOAT", "shape": [ "Addencoder_out_dim_0", "encoder_out_length", 25055 ] } ], "op_hist_top": { "Constant": 2405, "Add": 632, "Unsqueeze": 566, "Shape": 563, "Gather": 563, "Transpose": 490, "Reshape": 423, "MatMul": 421, "Concat": 352, "Cast": 222, "Mul": 214, "LayerNormalization": 142, "Split": 70, "ConstantOfShape": 70, "Slice": 70, "Pad": 70, "Conv": 70, "Softmax": 70, "Relu": 70, "Range": 3, "Sub": 2, "Div": 2, "ReduceMax": 1, "Less": 1, "Neg": 1, "Exp": 1, "Sin": 1, "Cos": 1 }, "dtype_hist": { "FLOAT": 233990207 }, "custom_domains": [], "has_external_data": false, "external_refs": [], "quantized": false, "model_type": "audio-speech", "file_size_mb": 893.76, "declared_weight_bytes": 935960828, "declared_weight_mb": 892.6, "weight_delta_mb": 1.16, "dtype_stats": { "FLOAT": { "count": 916, "bytes": 935960828, "mb": 892.6, "avg_bits_per_elem": 32.0 } }, "tensor_preview": [ { "index": 0, "name": "model.encoders0.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 1, "name": "model.encoders0.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 2, "name": "model.encoders0.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 3, "name": "model.encoders0.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 4, "name": "model.encoders0.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 5, "name": "model.encoders.0.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 6, "name": "model.encoders.0.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 7, "name": "model.encoders.0.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 8, "name": "model.encoders.0.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 9, "name": "model.encoders.0.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 10, "name": "model.encoders.1.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 11, "name": "model.encoders.1.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 12, "name": "model.encoders.1.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 13, "name": "model.encoders.1.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 14, "name": "model.encoders.1.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 15, "name": "model.encoders.2.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 16, "name": "model.encoders.2.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 17, "name": "model.encoders.2.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 18, "name": "model.encoders.2.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 19, "name": "model.encoders.2.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 20, "name": "model.encoders.3.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 21, "name": "model.encoders.3.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 22, "name": "model.encoders.3.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 23, "name": "model.encoders.3.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 24, "name": "model.encoders.3.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 25, "name": "model.encoders.4.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 26, "name": "model.encoders.4.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 27, "name": "model.encoders.4.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 28, "name": "model.encoders.4.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 29, "name": "model.encoders.4.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 30, "name": "model.encoders.5.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 31, "name": "model.encoders.5.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 32, "name": "model.encoders.5.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 33, "name": "model.encoders.5.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 34, "name": "model.encoders.5.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 35, "name": "model.encoders.6.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 36, "name": "model.encoders.6.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 37, "name": "model.encoders.6.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 38, "name": "model.encoders.6.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 39, "name": "model.encoders.6.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 40, "name": "model.encoders.7.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 41, "name": "model.encoders.7.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 42, "name": "model.encoders.7.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 43, "name": "model.encoders.7.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 44, "name": "model.encoders.7.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 45, "name": "model.encoders.8.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 46, "name": "model.encoders.8.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 47, "name": "model.encoders.8.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 48, "name": "model.encoders.8.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 49, "name": "model.encoders.8.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 } ] }, "tokenizer": {}, "tensors": { "count": 916, "preview": [ { "index": 0, "name": "model.encoders0.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 1, "name": "model.encoders0.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 2, "name": "model.encoders0.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 3, "name": "model.encoders0.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 4, "name": "model.encoders0.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 5, "name": "model.encoders.0.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 6, "name": "model.encoders.0.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 7, "name": "model.encoders.0.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 8, "name": "model.encoders.0.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 9, "name": "model.encoders.0.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 10, "name": "model.encoders.1.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 11, "name": "model.encoders.1.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 12, "name": "model.encoders.1.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 13, "name": "model.encoders.1.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 14, "name": "model.encoders.1.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 15, "name": "model.encoders.2.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 16, "name": "model.encoders.2.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 17, "name": "model.encoders.2.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 18, "name": "model.encoders.2.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 19, "name": "model.encoders.2.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 20, "name": "model.encoders.3.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 21, "name": "model.encoders.3.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 22, "name": "model.encoders.3.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 23, "name": "model.encoders.3.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 24, "name": "model.encoders.3.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 25, "name": "model.encoders.4.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 26, "name": "model.encoders.4.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 27, "name": "model.encoders.4.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 28, "name": "model.encoders.4.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 29, "name": "model.encoders.4.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 30, "name": "model.encoders.5.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 31, "name": "model.encoders.5.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 32, "name": "model.encoders.5.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 33, "name": "model.encoders.5.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 34, "name": "model.encoders.5.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 35, "name": "model.encoders.6.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 36, "name": "model.encoders.6.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 37, "name": "model.encoders.6.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 38, "name": "model.encoders.6.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 39, "name": "model.encoders.6.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 40, "name": "model.encoders.7.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 41, "name": "model.encoders.7.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 42, "name": "model.encoders.7.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 43, "name": "model.encoders.7.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 44, "name": "model.encoders.7.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 45, "name": "model.encoders.8.self_attn.linear_out.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 }, { "index": 46, "name": "model.encoders.8.self_attn.linear_q_k_v.bias", "dims": [ 1536 ], "dtype_name": "FLOAT", "byte_len": 6144, "bits_per_elem": 32.0 }, { "index": 47, "name": "model.encoders.8.self_attn.fsmn_block.weight", "dims": [ 512, 1, 11 ], "dtype_name": "FLOAT", "byte_len": 22528, "bits_per_elem": 32.0 }, { "index": 48, "name": "model.encoders.8.feed_forward.w_1.bias", "dims": [ 2048 ], "dtype_name": "FLOAT", "byte_len": 8192, "bits_per_elem": 32.0 }, { "index": 49, "name": "model.encoders.8.feed_forward.w_2.bias", "dims": [ 512 ], "dtype_name": "FLOAT", "byte_len": 2048, "bits_per_elem": 32.0 } ], "dtype_histogram": { "FLOAT": 233990207 }, "dtype_stats": { "FLOAT": { "count": 916, "bytes": 935960828, "mb": 892.6, "avg_bits_per_elem": 32.0 } }, "integrity": { "declared_weight_bytes": 935960828, "weights_size_est_bytes": 935960828, "weight_delta_mb": 1.16, "has_external_data": false, "external_refs": [] } }, "security": { "grade": "A", "score": 94, "findings": [ { "severity": "INFO", "category": "Shape Risk", "title": "Dynamic tensor dimensions detected", "detail": "Inputs with dynamic dims: ['speech']. Outputs: ['encoder_out'] (symbolic: Addencoder_out_dim_0, encoder_out_length, speech_length). Dynamic dims allow flexible batch sizes and sequence lengths, but without enforced bounds a caller can supply arbitrarily large tensors causing OOM. This is expected for most inference models.", "recommendation": "Enforce maximum input shape at the serving layer (e.g. ORT SessionOptions.add_session_config_entry, preprocessing validation). Document accepted input bounds in model card or deployment README." }, { "severity": "LOW", "category": "Supply Chain", "title": "No license field in ONNX model metadata", "detail": "Models without license information cannot be safely redistributed or deployed commercially. ONNX metadata supports a license property in model.metadata_props.", "recommendation": "Add a license field during export (e.g. apache-2.0, mit, cc-by-4.0)" }, { "severity": "LOW", "category": "Supply Chain", "title": "No provenance metadata in ONNX model", "detail": "Neither author nor domain (producer namespace) are set. Without origin information the model cannot be traced to a trusted source.", "recommendation": "Set author and domain fields during model export" }, { "severity": "INFO", "category": "Resources", "title": "Large ONNX graph — 7,496 graph nodes / ~234M parameters", "detail": "This model has 7,496 nodes, 916 initializers, 233,990,207 parameters, and 936.0 MB of declared weights. Larger graphs have higher memory allocation overhead at load time and may hit per-request limits in serving frameworks.", "recommendation": "Validate memory and latency under representative maximum input length. Set request concurrency and input-size limits before exposing the model as a service." }, { "severity": "INFO", "category": "Provenance", "title": "Model producer: pytorch 2.3.0", "detail": "Producer metadata helps identify the tool that exported this model", "recommendation": "Verify producer matches expected export pipeline" }, { "severity": "INFO", "category": "Classification", "title": "Inferred model type: audio-speech", "detail": "Based on op inventory (7496 total nodes, 28 unique types)", "recommendation": "Verify this matches the model's intended use case" }, { "severity": "INFO", "category": "Runtime", "title": "Default opset: 17", "detail": "Opset governs which ops and behaviors are available at runtime", "recommendation": "Use ONNX Runtime (CPU/CUDA/DirectML/OpenVINO backends); requires ORT ≥ 1.14" } ] }, "compatibility": [ { "runtime": "ONNX Runtime", "direct_supported": true, "conversion_possible": true, "verification_required": false, "quant": null, "arch": true, "context": "opset 17 (requires ORT ≥ 1.14)", "notes": "ONNX Runtime (CPU/CUDA/DirectML/OpenVINO backends); requires ORT ≥ 1.14" }, { "runtime": "TensorRT", "direct_supported": false, "conversion_possible": true, "verification_required": true, "quant": null, "arch": null, "context": "opset 17", "notes": "Engine build required (trtexec / TRT Python API)" }, { "runtime": "OpenVINO", "direct_supported": false, "conversion_possible": true, "verification_required": true, "quant": null, "arch": null, "context": "opset 17; conversion required", "notes": "Conversion via Model Optimizer (mo) or OpenVINO OVC (openvino.convert_model) required | Op coverage varies; validate with mo --input_model before deployment" }, { "runtime": "CoreML (Apple)", "direct_supported": false, "conversion_possible": false, "verification_required": true, "quant": null, "arch": null, "context": "opset 17; conversion via coremltools; opset out of range", "notes": "Conversion path via coremltools.converters.onnx.convert() — CoreML is not a direct ONNX runtime; the model must be converted to .mlpackage first. opset 17 exceeds CoreML converter limit (≤ 16). " }, { "runtime": "llama.cpp / Ollama", "direct_supported": false, "conversion_possible": false, "verification_required": false, "quant": false, "arch": null, "context": null, "not_supported_msg": "ONNX not supported — requires GGUF format", "notes": "GGUF format only; convert with llama.cpp convert scripts" } ], "report": { "summary": "ONNX audio-speech model | opset 17 | 7496 nodes | 233,990,207 params | 2 warnings / 5 informational", "bullets": [ "ONNX audio-speech model | opset 17 | 7496 nodes | 233,990,207 params | 2 warnings / 5 informational", "Size: 893.76 MB", "SHA256: b1145eb7e9d2dbb9..." ], "recommendations": [ "Add a license field during export (e.g. apache-2.0, mit, cc-by-4.0)", "Set author and domain fields during model export" ] } }