I wanted to build a poker solver that I could host for ~free. This meant running client-side, where the best browser technology is WebGPU. The tensor library I wanted did not exist, but in the LLM era we sometimes don't need libraries at all. I had Codex turn my thousands of lines of PyTorch code into custom WebGPU kernels with parity tests. Codex then spent days optimizing those kernels.

You can use the solver at https://holdem.computer, and the source is at https://github.com/phulin/poker2.

>>A more modern approach instead “re-solves” each spot to a limited search depth and uses a neural network as an approximation function at the depth cutoff. Both tabular (e.g. Piosolver) and neural (e.g. GTOWizard) commercial solvers are available.

PioSOLVER doesn't use any abstractions or cutoff functions. It just solves the whole game without any simplifications other than allowed bet sizes. The cost is rather large RAM requirements. The advantages is that it's very precise and produces exact results for every hand (it doesn't bundle them).