The linear no-threshold model or LNTM is a model of the damage caused by ionizing radiation, and particularly the increased risk of cancer.

It assumes that the response is linear and that this linear relationship continues to very small doses, that is to say that there is no threshold of exposure below which the response ceases to be linear. So, if a particular dose of radiation is found to produce one extra case of a type of cancer in every thousand people exposed, the LNTM predicts that one thousandth of this dose will produce one extra case in every million people so exposed, and that one millionth of this dose will produce one extra case in every billion people exposed.

A linear model has long been used in health physics to set maximum acceptable radiation exposures. In this application the question of a threshold does not arise, so this is not strictly the LNTM.

Some authorities regard the LNTM as extremely conservative or even completely wrong for predicting the effect of low doses of radiation. They claim that there is no evidence supporting the assumption that there is no threshold, and that recent studies contradict this assumption, see radiation hormesis. Others, and particularly opponents of nuclear power, support its use.