Multimodal AI issue – chat and image not working together

[crisp fern]

Hey, I’m Nate 👋

I’m building an AI project called Hymenoptera using Next.js and Vercel.

What works:

• The site is deployed
• Chat works
• Image generation works

What’s not working:

• I don’t know how to combine them into one system (multimodal)

What I expected:

• One AI that can handle both chat and images together

Project:
https://hymenoptera-ai.vercel.app
https://github.com/Topmonarch/Hymenoptera

I’m new to coding so even a small pointer would help 🙏

Multimodal AI issue – chat and image not working together

[tropic oxide]

Build an AI agent that combines those two functionalities

I did that for my text gen and image gen

Models

[crisp fern]

An AI agent, sure I can do that but if not, I'll just pay one of the freelancers to do it just in case if I can't Adjust the whole System and program it properly.

When I get that Finished, feel free to jump on my Hymenoptera, as soon as I pay for the domain

[magic patrol]

Hello, I can help you to create an AI agent that will connect everything together, If you don't mind

I will like to work and help you with this

[tropic oxide]

Same

[supple bone]

okay, I see

[fallow crown]

Hey @crisp fern I tried to register but this happned,

I tried it as guest, some things i find:

• there can be many optimizations on front end, like when i opened settings, then went back to chat ui did not changed, i have to actively looked for back button, I am choosing model type from top-bar but agent type from side-bar, Markdown o/p is not being rendered in md format but plane text, but apart from front end,

• I find all agents working in same, like all are able to code, reasearch
  and I don't know how to use business agent, how will it help me.
  I asked it how this agent can help me, it was not aware about agents it have, here is no persona to agents, output format, or guardrails

UI is clean though, its is more like bootstrap style,