Claude Fable 5.1 and GPT-6 Astra: Which Model Gets Which Job
AI News & Strategy Daily with Nate B. Jones
September 10, 2026
What changes when two AI models can turn the same short prompt into two different, usable apps? Nate compares Claude Fable 5.1 and GPT-6 Astra by building clipboard tools, trying them, and asking for the changes that only become clear after real use. How one opening prompt produced Ledge and Shelf.
Speakers Nate B. Jones
TopicsBusinessTechnology
Nate B. Jones (0:00)
I gave one prompt to two models, and I got two completely different apps, both trying to solve the exact same problem. Copy, paste, help. Fable built Ledge, a narrow list that slides in from the right side of my screen. Astra decided on Shelf, a wide bar across the bottom with nice big cards and a preview pane. They both got the exact same brief. They both got the same amount of time, and both of them are functional apps. They're good apps, and I'm a little bit surprised myself. But the thing nobody is measuring is how you compare two models in the second round. What if you get past that wide coding five line prompt? What do you do then? How do you go from there to done? That's what we're going to go through. It's the past you make after you've held the thing in your hands. And I think that matters more than ever now.
You can get both Shelf and Ledge at the end of this video. I'll put the link up and it'll be super easy for you. So context, I gave chat GPT-6 Astra and Claude Fable 5.1 the exact same job. I said, please, I struggle with copy and pasting. I've talked about copy pasting hell on this channel. Please, please, please build me a native Mac clipboard manager that feels like Apple should have included it. Something that remembers the text, the images, the links I've copied, lets me pull them back with a keyboard shortcut. I don't have to remember which one it was. I don't only use the most recent one, et cetera. So for this app experience, Astra got me something that I liked, that I actually use for copy and pasting. It got me there using fewer tokens. It saved me about an hour and it gave me more time to try it and ask for changes. So I got through more revolutions of the app. Fable gave me a design that I also liked, but ultimately it wasn't as usable because of tiny little details like hotkeys. And you'll see that as we dive in more. And I really, really care about what I can actually hand over and how I use what comes back. So I'm more productive with AI. And in this case, the job was small enough that I could try it immediately. Right? I could copy something. I could open up the panel in Ledge, open up the panel in Shelf. I could find it again. I could get back to what I was doing.
And I just got into the habit of actually using it. And that's the ultimate test for software. Now let's look at some of the different choices these models made. Ledge is a narrow list view that pops in from the right of the screen. You can copy a color. It will show you a whole swatch.
Code looks like code. Shelf doesn't have quite that precision. Shelf is wider. It has bigger cards to look at. It has filters and it has a larger preview. You can see really different ideas about how this tool would fit into your day-to-day. And that is one of the things I love, is that I can try both out functionally and evolve them as I go. I really like that Fable gave me something meaningfully different. When you ask both models for the same thing, one of the really useful outcomes is discovering that you hadn't decided exactly what you wanted, nor could you with a five-line prompt. Did I want a discrete list? I didn't make that call. Did I want more room to inspect an image? I hadn't decided. Seeing different design decisions well executed across two different form factors, those preferences are much easier to express and evolve and understand.
Both models received exactly the same opening brief, but they absolutely didn't take identical paths forward, and I am so glad they didn't. Fable decided to ask follow-up questions. It included which edge I wanted it on. Astra chose a bottom shelf on its own. I ended up using Fable through Claude Cowork and Astra through Codex on my Mac. So the experience includes the applications and access around the models, as well as the models themselves. And that's pretty typical with models and harnesses these days. Setup friction becomes real friction for someone who's trying to get work done very, very quickly. And what's interesting about both of these models and the apps that they created is that they illustrate how friction can feel different to different folks. I wouldn't want to call the distinct differences in these design decisions a difference in intelligence. Instead, I think what we're seeing is a distinct perspective that helped me as the app creator understand better what I was looking for out of that solution space. Shelf opens with Control-Shift-Spacebar. I love that big spacebar key. It was really easy as a hotkey to just grab it and go. And there was something intuitive about having the spacebar be at the bottom of my keyboard and the shelf pop up from the bottom.
13 more minutes of transcript below
Thousands of transcripts fetched by people building searchable podcast archives
Fetch the whole transcript
The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000788900665