would be great to be able to send streams and sequences to the api, when I try from LLM Vision I get an error
Error: ‘str’ object has no attribute ‘get’ as response.
When I switched LLM Vision to use OpenAI’s gpt-4o-mini it worked fine with the same call.