I'm assuming that you've watched a recording, or read a write-up, of a talk I gave in 2020 where I was explaining how to run TensorFlow models in Scratch - and you zoomed in close to see the URL of my model hoping to reuse it for yourself?
If that's correct, I'm afraid you misunderstood what I was saying. I was showing how to do it, and explaining what was involved. I wasn't intending to host models on my personal server (and in fact, I deleted the model you saw me demonstrate immediately after giving that talk). I don't think you're missing out on much - it was a quick toy model I was using as a demonstration of the steps involved - I don't know why you'd want that model in particular.
Stepping back, the talk had three sections:
1) Using your own TensorFlow models in Scratch
- this is where you train your own model
2) Using pretrained models in Scratch
- this is where you find well-known models in central repositories. I used TensorFlow Hub for this. That's a sign of this being a nearly-six-year-old presentation, as that site doesn't exist any more, having been largely replaced by Kaggle.
3) Creating TensorFlow models with Teachable Machine and using them in Scratch
The point of the talk was to say that you don't have to be restricted to the models that I host for you on Machine Learning for Kids - you can create your own, or go and choose from the many thousands of models available on the Internet.
With the exception of suggesting TensorFlow Hub, I think the rest of that talk still is largely accurate (unusual for such an old talk about AI!) so it might be worth going back and having another look.
Kind regards
D