Started 2 workflows to use Nano Banana to combine multiple images into one image, but have had a roadblock. n8n does not handle multiple binaries at the same time very well. I can split them out but converting the binaries into base64 does not go well. I'm sure there's a way to do it but I have not had much luck with more than an hour of trial and error, ChatGPT and Perplexity searches. I'll have another video about Nano Banana soon, but anybody else run into issues using multiple binaries in the same flow and trying to convert those binaries to other file formats? Let me know if you have any tricks!

We were playing around with Google Opal earlier this week and it's a promising tool. With one prompt, I had a workflow that would combine multiple images using Google's new Nano Banana image model. Opal is limited to Google tools only at this point in terms of input and output, but Google has an impressive suite of abilities so if you're already invested in the Google ecosystem, this could be an easy and fast way to combine your workflows without having to connect with webhooks and triggers. Plus one of the output options is a custom Webpage so you're well on your way to a custom SaaS. You can try it out for free right now at https://opal.withgoogle.com/

What's the plan for this week? One big goal for Friday? I've got my eye on an ElevenLabs alternative, whether it's Microsoft's new VibeVoice model or Kokoro (thanks for the suggestion @Danilo Alves!), a self-hosted voice system will be in the works. I'll have a video posted by the end of the week. And maybe some bonus nano-banana since that seems to be a popular one. Let me know what else you'd like to see or need help with. Happy Monday and good luck this week!

Google's new Gemini-2.5-Flash-Image-Preview model, codenamed "nano banana" before launch dropped yesterday on aistudio.google.com. It was being called the "Photoshop killer" since it can very accurately follow your natural language instructions to edit images. Plus it's not bad at generating images either. Planning to have a workflow to demonstrate the power of editing here. I'll have the video up soon. Have you tried using the model yet?

As I use them more, it’s becoming more obvious what AI models are good at what, which is partly why I don’t like the “Auto” setting in Cursor, and partly why the GPT-5 launch had so much backlash. It seems like GPT and Claude models tend to rely more on their training. Perplexity relies more on searching the web for resources. AI training can go out of date pretty quickly, or it can be averaged into a wrong answer. Search results are more up to date and reliable. So generating code is fine for Cursor models, it just needs to be pretty close syntax averages to produce working code. But asking about best practice can be an issue if that process has changed, or you need something specific. Perplexity will find the docs and reference those, while other models will default to their training and generate an answer that isn’t a direct reference, like trying to recall from memory, not read the paper directly. If anything, that’s the easiest distinction. For up to date references, Perplexity is probably the best answer. It’s entire purpose is search. Grok does prioritize X.com posts and web posts so it does a decent job of generating a current answer as well. Gemini with the “Grounding with Google Search” option enabled seems pretty good in making sure it’s answers reflect real documents too. GPT and Claude are good at calling tools, but seem to rely more on generating answers as opposed to referencing answers. GPT does have 2 different options in the browser and app that make web search the priority, but in an API format like Cursor or n8n, that is not enabled by default I’m sure. It has to be expressly prompted to search, which makes sense, internet search is not cheap. So moral of the story, if you’re looking for answers on up-to-date best practice, Perplexity is probably the first choice. If you're looking for up-to-date research Perplexity is also the first choice. I have a research workflow on the list I'll be sure to demo the Perplexity connection or MCP even in the near future!