Objective. Many studies find that females benefit from their gender in sentencing decisions. Few researchers, however, address whether the gender-sentencing association might be stronger for some crimes, such as minor nonviolent offending, and weaker for other offenses, such as serious violent crime. Method. Using a large random sample of convicted offenders in Texas drawn from a statewide project on sentencing practices mandated by the 73rd Texas Legislature, logistic regression and OLS regression analyses of likelihood of imprisonment and prison length illustrate the importance of looking at sentencing outcomes not only in terms of gender but also in terms of crime type. Results. Specifically, we find that the effect of gender on sentencing does vary by crime type, but not in a consistent or predicted fashion. For both property and drug offending, females are less likely to be sentenced to prison and also receive shorter sentences if they are sentenced to prison. For violent offending, however, females are no less likely than males to receive prison time, but for those who do, females receive substantially shorter sentences than males. Conclusions. We conclude that such variation in the gender-sentencing association across crime type is largely due to features of Texas’ legal code that channel the level of discretion available to judges depending on crime type and whether incarceration likelihood or sentence length is examined.