The invention relates to a length-controllable text abstract generation system and method. The length-controllable text abstract generation method comprises the following steps: 1, preprocessing a source text submitted by a user and length control information to obtain input data capable of being processed by a text abstract model; 2, on the basis of a deep learning generative model, combining a text abstract system with a copying mechanism, a coverage rate mechanism and a length control mechanism to process and calculate the input data to obtain an internal representation of the generated abstract; and 3, based on the text internal representation generated in the step 2, further converting the text internal representation into an abstract result with controllable length, and presenting the result to a user by the system and visually presenting the result. On the basis of the deep learning generative model, the copying mechanism, the coverage rate mechanism and the length control mechanism are added, thus achieving abstract generation of the long text, and being high in readability and controllability, and an evaluation result shows that the abstract generation effect of the methodis obviously improved compared with that of a standard generative model.