The aim of this study was a performance comparison of two clean-up procedures (dilutions versus immunoaffinity columns) in the simultaneous determination of eight mycotoxins (aflatoxin B1, deoxynivalenol, fumonisin B1 & B2, ochratoxin A, toxin T-2 & HT-2 and zearalenone) in the animal feed. After extraction the analytes were separated on a Kinetex Biphenyl column with a gradient elution using methanol/0.01 M ammonium acetate as a mobile phase and analyzed with the LC-MS/MS technique. Both of the procedures were validated by analysis of a series of spiked feed samples (n = 6) at three different concentration levels. Better signal to noise ratios were observed for immunoaffinity clean-up. The recoveries of analyses were in the range 88-110% for the dilution procedure and 78-120% for the immunoaffinity clean-up. The dilution procedure was more precise (coefficient of variation of the within-laboratory reproducibility for it was 7.8-22.4% in comparison to 12-35.5% for the immunoaffinity clean-up. The results show that both procedures fulfilled the requirements for mycotoxin analysis and can be used successfully in multi-analyte determination. Although the dilution procedure shows better precision and trueness, the immunoaffinity clean-up procedure can have advantages in more complex feed samples thanks to lower matrix effect and limits of detections.