Pinned Tweet
Update on the project comparison (created a "timekeeping" app attempted in a simultaneous one-shot session with Claude Code Opus4.6 and GPT5.4): I hope this is helpful. I think immediately a clear winner emerged. Give feedback and let me know if I left something out. I gave the exact instructions with identical plans (created by Grok for fairness) to both models. Here's what happened:
-Both finished the first pass in about the same time, within a few minutes about 40 min total. GPT produced in stages so a little more time 20 min was required notes below.
-GPT produced a "initial MVP" purely for viewing and did this several times. Rather than delivering every step at once, GPT allowed confirmation at several steps. I like this approach. Catch errors in stages. Took extra time, but it was welcomed.
-The first pass was more complete from Claude, but not everything worked refactoring required on most of the windows, especially errors involving CRUD operations. I had to request refactoring several times, still trying to resolve CRUD issues. Not resolved.
-I actually liked the frontend GPT5.4 generated, it's actually very pleasing to use.
-Claude had several parts that did not include add, edit or delete options. I could not add team or clients. After the "Add" buttons were included, they did not work.
-Tried to enter a new client in Claude, it would not add. It froze, then the record did not add, same for projects.
-GPT5.4 is building everything in steps and requests checks as we go along, I kind of like this feature. So far everything is working but the next phase is the CRUD, where problems are usually found. Essentially no errors.
Summary with winners (sorry not in a table):
Frontend design: GPT
Fully functioning: GPT (after stepping through phased output), everything worked no issues
CRUD: GPT, Claude is still not ready, taking several passes and still does not work. I'm not going to keep trying.
Fields included: GPT was much more thorough with all the fields as specified in the design plan. Claude took the minimal approach, left fields out.
Followed the design: GPT, I'm still checking this, but appears every design element was followed. Claude omitted many items.
Code review: Tie - this was done at a high level but the code seemed cogent without strange patterns, linting was self checked by both, and no problems were found (not the best analysis, but fine for purposes here).
Token use: Claude 45k, GPT recorded zero tokens, this has to be configured up front and I didn't do this. Apologies on this, as mentioned I have the max subscriptions on everything so I usually don't pay attention to the costs, unless I hit a wall, which we rarely do.
Winner analysis
Here's my assessment. GPT5.4 clearly won this one. It followed the design, gave steps for checking and after adding CRUD operations everything worked without error. Plus the frontend design is so pleasing and easy to work with in GPT. I would actually enjoy working in the GPT app and I might end up using this internally. Not sure about making it a product offering though. I'm a huge general Claude fan, I use it for many things, but Claude Code (CLI) needs work. I think Opus 4.6 is smart enough, but for some reason it does not produce rock-solid out of the gate. This has happened with several projects.
Screen shots:
Claude's first pass (the plain looking one), and GPT's first pass (the fancier looking window); followed by GPTs project entry window followed by Claude project entry window. I'm not sure how the images will show, but GPTs windows have much more verbiage and has the very light tan background. Claude images are black and white.
Perhaps an unpopular opinion but: I really like the output that ChatGPT/Codex creates, but the ChatGPT/Codex app itself is very difficult to keep organized. Chats go to work objects, work creates chats, the "Spaces" component creates items all over the place, work sometimes, I try to isolate project work, but it creates stuff everywhere. My team spends half of their time asking, "where did it go?"
We have returned to Claude for development of some projects recently. We still use Codex/ChatGPT, but the latest Claude app works very well. I always thought the Opus models were good, but the harnessing seemed a little haphazard. That is no longer the case. Using Opus 5.5 Medium and it's plenty "smart" enough.
Wow. GPT5.6Sol just created an email response that sounds exactly like how I would respond!
A customer sent a tome of an email with concerns and requirements. To thoroughly respond I let GPT5.6Sol check it and write a response. Amazing! It created an email that was casual language (like mine) with not a single em dash. Now I wonder if the model looked at my prompts and tried to mimic my style. Interesting thought.
I'm enjoying building with Grok 4.5. I'm not a huge fan of the CLI but it does have some advantages. I'm looking forward to a desktop app for Grok Build.
Codex has had a bunch of big updates in the last few days. I hope they are good changes. Evaluations underway.
Building with Fable 5 (leaving us July 12), its ability seems slightly higher than Opus 4.8, and more "controlled" in its code creation. Return messages are much more complex and brief. It still makes assumptions without confirming, and they are not correct, requiring some rework. Also my computer is smoking from the token usage. I still lean towards Codex and Grok Build just a little bit. They are slightly slower, but they don't make assumptions without confirming in a proposed plan first.
Building a new project using Fable 5. Honestly, it's difficult to determine if it's "smarter" than the Opus models, but it seems much more "in control". The problem we had with Opus models in the past, is that it would make sweeping changes in code without confirming that it was a good plan. Several times, that created hours of corrections. Fable doesn't seem to have this problem. That's very good progress!
Grok Build just helped me create an ERP addin for Odoo. Flawless. No errors at all (which is not easy to do) and works perfectly. I'm duly impressed. I am now wanting more of it. The CLI works fine, but I do like having a standard desktop app. I hope that is available soon! Maybe it is already?
Up super early on a Sunday morning coding. It's a sickness I guess. Using Grok Build on a project. The CLI product is getting better with updates. I think this will be a viable product!
Experimenting with Grok Build. The updated version of is much tighter and more organized and a little less repetitive. I'm enjoying its approach in coding! And it's good at UI.
I've been building a new project with Grok Build and I've enjoyed the experience. The CLI is not bad, the code development is good, the actual output is very good. The UI design is good, but uses a bunch of tiles and builds too many functions on a page (uni-tasker windows work best for most scenarios). Most of the models I work with do the same thing, though, if left to their own design planning. Somewhere someone must have posted a blog saying "use lot of tiles and add as many functions to a page as possible" 🙂
Grok build seems to be on the right track. I've been using it to build a new app, and it's quite good. The token usage is not bad either. It does need a $100/month rate though. Right now the rates are $30/month, then $300 per month. There is a "pay as you go" feature. So far the $30 per month is lasting quite a while (this is an official dev measurement of time).
Grok Build is working nicely! It's worth a try for sure! There are some minor needed refinements, but this is a beta version.
It repeats responses quite a bit (minor), but it follows directions very effectively. It remembers I randomly requested a long time before.
Give it a try and good luck!
Up early as always, coding. I'm actually enjoying Grok Build CLI. It's faster than GP5.5 (which is not always good), but it's UI capabilities and planning capabilities are very good. I think this will become a useful tool!
I'm testing Grok Build with a new project. I'm not a huge fan of CLI, but it's powerful and in this case Grok is doing a great job so far. Super fast at this point. I'll report more as progress is made.
So a very large customer says "hey we cannot find this functionality out there, can you build it???". Of course I said "yes". How could I say "no" to my new project addiction. But this one already has a need!
Building a nice specialty software for scheduling/timekeeping for first responders using GPT5.5 (med). The model has performed perfectly.
I've asked the team to try to estimate the time savings and they estimate about 75% to 80% time savings for coding, maybe 25% unit testing time savings. No real savings in code review and UI design. The last two parts cannot be omitted.