Heads-up no-limit Texas Hold'em research stack with exact Python/C++ engines, Deep CFR training, evaluation, search, and a desktop GUI.
High-throughput reinforcement-learning search and sampling in Rust, with caller-owned Python networks and training
Implementation of AI Poker HULH with Deep CFR and MCCFR ES as thesis for Bachelor of Engineering degree
6-max NLHE poker bot on Deep CFR. MCCFR-trained, pure-NumPy runtime, 2s/768MB sandbox. 12th of 500+, 15th of 64 in the final, Fullhouse Hackathon 2026 (Quadrature Capital).
Research prototype for 6-max poker AI systems using Rust, Deep-CFR-style training, PyTorch, ONNX, and controlled simulator evaluation.
Open-source GTO poker solver with neural network strategy approximation and 75ms real-time search decisions. Built for study, research, and extension.AI built in five stages: CFR, card abstraction, Deep CFR, and real-time search. Readable code, full docs, 27 tests.
To associate your repository with the deep-cfr topic, visit your repo's landing page and select "manage topics."