Multimodal discussion for Gemma

[torchshed66]

Posting this in case it saves someone else a little time.

If you were starting from scratch today for this use case, what would you choose first and what would you ignore?

I'm especially interested in mistakes worth avoiding.

[macropilot]

I hit almost the same issue last night.

[vramport]

I hit almost the same issue last night.

Good point. I ended up in about the same place using TensorRT-LLM. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.