.Net: Bug: Using WithOpenAITextEmbeddingGeneration() with locally hosted model gives error "The input is not a valid Base64 string of encoded floats."

[ChiragSoni95]

Describe the bug
using WithOpenAITextEmbeddingGeneration(), I get the error "The input is not a valid Base64 string of encoded floats." in semantic kernel gives "The input is not a valid Base64 string of encoded floats."

To Reproduce
Steps to reproduce the behavior:

• See usage and error screenshots below

Expected behavior
Successfully get embedding and add to the MemoryStore

Screenshots

Platform

• OS: Windows
• IDE: VSCode
• Language: C#

Additional context
I am using LMStudio for hosting a local embedding model, the precision value of floating point is different than what AzureOpenAIAPI has, see below: is that causing issue using above code?

AzureOpenAI API Embedding Output: This works fine with volatile memory store

{"object": "embedding", "index": 0, "embedding": [ -0.012779145, -0.0075187064, -0.017534792, -0.028241644, -0.018796766,
...
}

LM Studio Embedding model output:
{ "object": "embedding", "embedding": [ 0.04841502383351326, 0.026464618742465973, -0.17441967129707336, -0.03099326603114605,
...
}

Format of the output JSON is exactly same except the precision of floating points.