Any workarounds for function calling using gpt-4-vision-preview?

[merefield]

See update here:

Does the model `gpt-4-vision-preview` have function calling?

So I gave up and wrote a function. With hindsight this has one huge advantage: you can use a cheaper LLM to hold the general conversation and restrict calls to GPT 4 Vision for just image analysis. UPDATE: looks like Open AI released an updated preview model with vision with function calling: gpt-4-1106-vision-preview (but as usual the alias gpt-4-vision-preview points to this latest model) More here: https://platform.openai.com/docs/models/continuous-model-upgrades My solution is stil…