{% extends "_base.html" %} {% block content %}
Pick a Hugging Face model, choose a target average bits-per-weight, and OptIQ picks per-layer precision from a sensitivity sweep.
{# Step indicator #} · · proceed with caution
Couldn't read config.json — model may be private or non-standard
n_layers × n_bits × n_samples forwards. 8 is the default (good signal/cost balance); 16+ shaves variance on noisy layers but doubles the wait.OptiQ-4bit) following llama.cpp / Unsloth convention.OptIQ quant saved to:
Uses the token from Settings → Hugging Face. Requires your Lab password to decrypt.