cpu-training

MasterAI 1.0 - 1v1 Limit Texas Hold'em AI /德州AI / 一对一限注德州扑克AI / 一對多限注德州撲克AI - CFR solver, Monte Carlo, real-time decision engine

RL GRPO Finetuning on CPU

CPU-trained reasoning model pipeline. LoRA SFT + DPO on SmolLM2-360M, GSM8K math reasoning, single-laptop deployment.

Local-first AI workspace with NP-DNA — a NeuroPlastic DNA Network for CPU-native training, memory, automation, and dashboard.

Interactive Training Dashboard & CAGS-Operator Verification for JamOne Nano.

Teaching a machine to learn code the way humans do — curriculum-based AI training on a laptop, no GPU, no cloud, no massive datasets. Proving that thoughtful data design beats brute-force scale.

An end-to-end pipeline for training and deploying a lightweight math reasoning language model (Qwen2.5-0.5B). Features CPU-compatible Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and an interactive web interface built with Flask and Streamlit.

Native ARM64 PyTorch on Windows on ARM (Snapdragon X Elite): the official wheel benchmarked and found slower than x86 emulation, plus the WSL2 path that actually trains. One-command emulation diagnostic included.

Fine-tunes Gemma-3-1b Math SFT on a custom JSONL dataset in <20min on CPU. Computes before/after eval.

34.1M LM on a laptop CPU: 215,771 tok/s, 8,529x Qwen. QSELM beats Qwen3.5-0.8B on sealed long-document QA (90.6% vs 45.8%) and Qwen3-0.6B-FC on cross-turn agent memory (69.6% vs 3.2%). No GPU. | 笔记本CPU训练34.1M模型：21.6万token/s，实测为Qwen的8,529倍；长文档问答，QSELM 90.6%，Qwen3.5-0.8B 45.8%；Agent跨轮次记忆，QSELM 69.6%，Qwen3-0.6B-FC 3.2%；无需GPU。