You need to agree to share your Hugging Face information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

mammoth-1

A typed decision model: a full fine-tune of Qwen/Qwen3.5-27B that answers every question of a request with a probability per option from one shared-prefix prefill, without generating text.

Run the Decision Index

On one CUDA GPU with pytorch/pytorch:2.8.0-cuda12.8-cudnn9-devel:

git clone https://huggingface.co/baselabs/mammoth-1 && cd mammoth-1
bash setup.sh /tmp/w
python -m decision_index pipeline --engine engine:DecisionModel --option model=. --option device=cuda:0 \
    --suite-dir <suite-0.3> --out runs/mammoth-1

setup.sh installs the runtime and the Decision Index kit, pinned by constraints.txt. engine.py is the kit engine.

Downloads last month: 14

Safetensors

Model size

27B params

Tensor type

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for baselabs/mammoth-1

Base model

Qwen/Qwen3.5-27B

Finetuned

(338)

this model