Most research on dyslexia to date has focused on early childhood, while comparatively little is known about the nature of dyslexia in adolescence. The current study had two objectives. The first was to investigate the relative contributions of several cognitive and linguistic factors to connected-text oral reading fluency in a sample of adolescents with dyslexia (n = 77). The second was to test the hypothesis that the effect of verbal working memory on connected-text oral reading fluency is moderated by word-level skills and/or vocabulary knowledge. The results suggest that many deficits associated with childhood dyslexia remain prominent in adolescence, but the nature of the relationships between key cognitive and linguistic predictors (i.e., word-level reading, vocabulary, verbal working memory) and reading fluency appear to be different in adolescence. For example, while word-level skills remain a significant predictor, the strength of the effect is relatively weak. In contrast, the data support an increased role for vocabulary and verbal working memory, including an interaction between these factors. The presence of an interaction can be interpreted as evidence that the influence of verbal working memory on connected-text oral reading fluency in adolescents with dyslexia depends on individual differences in vocabulary knowledge. These results offer support for the changing nature of dyslexia across development, and suggest that researchers should study dyslexia in adolescents on its own terms, rather than treating it as an extension of reading problems in early childhood.