The aim of this study was to determine the performance of real-time amplification based methods - NASBA, TaqMan, RT-FRET, and RT-PCR LUXtrade mark formats - for the detection of influenza A (H5N1) virus RNA. In an analysis of 54 samples obtained from a range of animal species in Thailand during the period 2003-2006, results showed that the NASBA (H5=98.2%, N1=96.3%), TaqMan (H5=98.2%, N1=96.3%) and FRET (H5=98.2%, N1=96.3%) had significantly higher rates of positive detection than LUX (H5=94.4%, N1=50.0%; P<0.001) for influenza A, H5 and N1 isolates. There were no false-positive results from any methods used in the negative-control group of samples. The limits of analytical detection were at least 10copies/reaction in real-time NASBA and LUX assays, while FRET and TaqMan assay appeared to be less sensitive at > or =100copies/reaction. The assays were relatively specific without cross-reactivity to a number of other influenza strains or viral pathogens. In conclusion, our study demonstrated that real-time NASBA, TaqMan and FRET assays can be used to detect influenza A (H5N1) from a wide range of hosts, and be specific for H5N1 samples obtained during different outbreaks (2003-2006). All assays provided the benefit of rapid influenza H5N1 identification for early diagnosis, in the range of hours, and they are well suited to high throughput analyses.