[gpjanik]

The real performance is nowhere close to what is presented in the marketing materials, which is pretty annoying. Especially text rendering and accuracy.

Try asking it for a plot of Polish GDP growth over the past 20 years. It's slop.

[viraptor]

This model isn't supposed to contain all the numerical data. It will give you a (usually) matching graph transformed from one you provide or from a table of information you provide. Or you can pipeline from an LLM doing research on that data first. But expecting an image gen model to get you GDP info has got to be one of the worst possible approaches.

> Especially text rendering

That's true though. I still got some completely fried letters in headings.

[yorwba]

Giving it numerical data in an LLM-generated prompt doesn't seem to help much: https://imgur.com/a/KFhczOd

It included the table verbatim and even managed to hallucinate a reasonable heading for it, but then the graph doesn't even manage to align the data points with the time axis, leading to an unfortunate collision in the middle.

I guess you should use a traditional graphing library for your presentation slides for now.

[spwa4]

... but if you want an accurate graph, why not ask the LLM model to put the data points into a graphing library?

[gpjanik]

I am not expecting it, it's the Qwen team is claimingthey can do much harder tasks than this, like rendering a consistent page of a maths paper, or creating true to fact explainers.

They can't.