Abstract Background Rapid diagnostic tests (RDTs) have become an essential tool in the contemporary malaria control and management programmes in the world. This study aims to evaluate the performance of two commonly used RDTs for malaria diagnosis in the China-Myanmar border area. Methods A total 606 febrile patients in the China-Myanmar border were recruited to this study and were diagnosed for malaria infections by microscopy, two RDTs tests (Pf/Pan device, and Pv/Pf device) and nested PCR. Results Malaria parasites were found in 143 patients by microscopy, of which 51, 73, and 19 were Plasmodium falciparum, Plasmodium vivax and P. falciparum/P. vivax mixed infections, respectively. Compared to microscopy, the sensitivity of the Pf/Pan device was 88.6% for P. falciparum and 69.9% for P. vivax with the specificity of 90.4%. For a subset of 350 patients, the sensitivity of the Pf/Pan device and Pv/Pf device for detection of P. falciparum was 87.5% and 91.7%, respectively; and for detection of P. vivax was 72.0% and 73.8%, respectively. The specificity of the Pf/Pan device and Pv/Pf device was 94.3% and 96.5%, respectively. Nested PCR detected malaria parasites in 174 of 606 samples, of which 67, 79, two and 26 were P. falciparum, P. vivax, P. ovale and P. falciparum/P. vivax mixed infections, respectively. Compared to nested PCR, all other methods had sensitivity below 80%, suggesting that a significant number of cases were missed. Conclusions Compared to PCR, both microscopy and RDTs had lower sensitivities. RDTs had similar performance to microscopy for P. falciparum diagnosis, but performed worse for P. vivax diagnosis. Other RDT products should be selected with higher sensitivity (and good specificity) for both P. falciparum and P. vivax diagnosis.