Skip to content
Other

Dualist-Hyper-v2

hybrid AI engine for Othello (Reversi)

2 min readupdated

Dualist Hyper combines a 20M-parameter Dual-Stream FiLM-SE-ResNet model with a C++20 AVX2 bitboard engine and an SIMD table. It tackles strong algorithmic and neural move planning for the board game Othello. Under the hood, a hybrid architecture processes spatial and scalar data streams alongside an Egaroucid-based search engine. The project is built for game AI developers, neural heuristic researchers, and board game enthusiasts.

Key features

  • Hybrid architecture: pairs a 16-block ResNet with a C++20 bridge built on the Egaroucid library.
  • Accurate predictions: outputs move policy logits, disc differentials, and win/loss distributions.
  • Efficient search: uses NegaScout, Multi-ProbCut, and an endgame solver for empty squares.
  • Strong gameplay: reaches a 55% win rate against Egaroucid Level 11 and 100% against Level 1 without opening books.

What you can do

  • Build a bot: integrate the engine to power a strong AI opponent in Othello games and bots.
  • Analyze games: run deep position searches and evaluations up to depth 13.
  • Run from Python: load model weights from model.safetensors and control the C++ engine via bindings in bridge_binding.py.

Example

from safetensors.torch import load_file
from modeling_dualist import DualistNetHyper
import json

cfg = json.load(open('config.json', encoding='utf-8'))['neural_architecture']
model = DualistNetHyper(
    in_planes=cfg['in_planes'],
    scalar_dim=cfg['scalar_dim'],
    channels=cfg['channels'],
    num_blocks=cfg['num_blocks'],
).eval()
model.load_state_dict(load_file('model.safetensors', device='cpu'), strict=True)
Read moreDetails
Benchmarks: 55.00% win rate against Egaroucid Lv11 (~2095 Elo); 18/20 optimal solutions on the FFO Endgame Suite.
Requirements: CPU with AVX2/BMI2 support, CUDA-capable GPU for GPU inference.
Limitations: GPL-3.0 license due to Egaroucid components.
Training details: trained on 88,901,875 game states using Sparse AdaGrad.
Discuss Telegram 0 19

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 0.1 GB · ready files0.6 GB
  • BF16weights 0.0 GB0.5 GB
  • FP8 / INT8recommendedweights 0.0 GB0.5 GB
  • 4-bitweights 0.0 GB0.5 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · F32 — from RTX 2060

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # brandonlanexyz/Dualist-Hyper-v2
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download brandonlanexyz/Dualist-Hyper-v2 --local-dir ~/models/Dualist-Hyper-v2
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to Dualist-Hyper-v2
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of Dualist-Hyper-v2 (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

brandonlanexyz/Dualist-Hyper-v2
Author
brandonlanexyz
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

190

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments

Other

Perth ONNX

Perth ONNX