This paper concerns the mode dependent H_{∞} filter design for continuous Markov jump linear systems. The filter gain to be designed is assumed to have additive variations and the transition probabilities are allowed to be known, uncertain with known bounds and unknown. Attention is focused on the design of a mode dependent nonfragile full order filter, which guarantees the filtering error system to be stochastically stable and has a prescribed H_{∞} disturbance attenuation performance. Sufficient conditions for the desired filter design are given in the framework of linear matrix inequality. If the filter gain variations become zero and the transition probabilities are completely known, the proposed method is reduced to the standard H_{∞} filtering results. A numerical examples is given to show the effectiveness of the proposed method.