GPT-5.6 Sol Ultra: $2,000 in tokens, 437,000 lines of AI-written code in one hour.

This is basically the current state of AI coding when you put it in the hands of an actual software engineer. I had it build a display server and compositor around a minimal Wayland-like design, but using a ring-style protocol. It benchmarks around 10%+ faster across the board, with some areas doing considerably better, and uses Vulkan instead of GLES.

From there, it built a 2D graphics engine, GUI toolkit, Flexbox layout engine, font and PNG rendering, a 3D graphics engine, and an AV1 decoder. Most of the rendering is Vulkan-based, with CPU rendering used where it makes more sense. 437,000 LOC of C++ 1hr time.


This is how I make sure it actually understands what it’s building. I first develop the concept or algorithm, help it implement it in C++, and then have it write a short document explaining how the algorithm works. In this case, I take that documentation, give it to a fresh AI, and have it implement the same algorithm in a different programming language. 300+ LOC of Crystal Lang 2min time.


Neither the design for the compositor nor the smart background erase brush exists in the wild. In fact, the background erase brush only exists in Photoshop and Affinity Photo. I couldn’t find any other image editor that has the same feature.


The 437K LOC only had 5 or 6 compile errors, no warnings, and the code is perfectly understandable and well commented. The 300+ LOC had no compile errors or warnings at all. In my opinion, the code quality of both is at the highest level.
 
I'm not a big fan of AI and vibe coding. I wanna learn coding for fun, not to push out some software for a corporation or development team every few months.
 
Back
Top