Generalist decision model (REAX SN92)

A LoRA fine-tune of perplexity-ai/pplx-decider-v1.2 (Apache-2.0), merged into the weights, for the REAX subnet's generalist decision track. Same protocol as the base: joint inline-readout v3, readout.safetensors + decision_config.json.

Training data includes items from decisionmodels/sn92-specialists (CC-BY-4.0, by the REAX subnet owner), permissively licensed public benchmarks' train splits, and decision items written by open-weight models.