### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

Including encoding_format as a list within model parameters leads to a vLLM failure on model calls that don't explicitly set a specific encoding_format, as by default LiteLLM passes the whole list to vLLM as a string.

``` "encoding_format": [ "float", "base64", "ubyte", "int8" ]

```

The model Test Connection behaves in this default way, and displays this error

`Error testing connection: litellm.BadRequestError: Hosted_vllmException - {"error":{"message":"3 validation errors:\n {'type'... `

Using the Swagger API you can reproduce this error by not setting encoding_format or otherwise setting it.

### Steps to Reproduce

1. Set encoding_format to a list on an embedding model 2. Make a call to the embeddings endpoint without specifying an encoding_format 3. Make a call to the embeddings endpoint with specifying an encoding format

### Relevant log output

```shell

```

### What part of LiteLLM is this about?

Proxy

### What LiteLLM version are you on ?

v1.83.14

### Twitter / LinkedIn details

_No response_