Aspects of the disclosure provide methods and apparatuses for video encoding/decoding. For example, processing circuitry decodes prediction information of a current block in a current picture from a coded video bitstream. The prediction information is indicative of an affine merge mode with offset. Then, the processing circuitry decodes, from the coded video bitstream, a set of offset parameters that is used to determine a motion vector difference, and applies the motion vector difference to first motion vectors of multiple control points of a base predictor of the current block to determine second motion vectors at corresponding multiple control points of the current block. Further, the processing circuitry determines parameters of an affine model based on the second motion vectors at the corresponding multiple control points of the current block, and reconstructs at least a sample of the current block according to the affine model.