Not able to load ONNX Transformer model (though successfuly used with DJL)

[JirHr]

I am trying to use ONNX export of model intfloat/multilingual-e5-large: https://huggingface.co/intfloat/multilingual-e5-large/tree/main/onnx

I have successfully downloaded this model, using it successfuly with Deep Java Library (ai.djl:api:0.28.0) - therefore I am confident, that model itself is correct.

My code snippet:
TransformersEmbeddingModel embeddingModel = new TransformersEmbeddingModel();
embeddingModel.setTokenizerResource("classpath:/onnx/multilingual-e5-large/tokenizer.json");
embeddingModel.setModelResource("classpath:/onnx/multilingual-e5-large/model.onnx_data");
embeddingModel.afterPropertiesSet();

embeddingModel.afterPropertiesSet() gives me following Error:
Caused by: java.lang.OutOfMemoryError: Required array size too large
at java.base/java.io.InputStream.readNBytes(InputStream.java:420) ~[na:na]
at java.base/java.io.InputStream.readAllBytes(InputStream.java:349) ~[na:na]
at org.springframework.util.FileCopyUtils.copyToByteArray(FileCopyUtils.java:149) ~[spring-core-6.1.10.jar:6.1.10]
at org.springframework.core.io.Resource.getContentAsByteArray(Resource.java:151) ~[spring-core-6.1.10.jar:6.1.10]
at org.springframework.ai.transformers.TransformersEmbeddingModel.afterPropertiesSet(TransformersEmbeddingModel.java:193) ~[spring-ai-transformers-1.0.0-M1.jar:1.0.0-M1]

The model has 2,1Gb, I am using 64bit JVM 17 and setting JVM -Xms16g -Xmx16g did not help.....

The model has two files model.onnx and model.onnx_data. Initially I was using model.onnx, getting ai.onnxruntime load error.
I have 32Gb memory, which should be fully sufficient for this model...