The VITEK 2 is a new automated instrument for rapid organism identification and susceptibility testing. It has the capability of performing rapid susceptibility testing of Streptococcus pneumoniae with specially configured cards that contain enriched growth medium and antimicrobial agents relevant for this organism. The present study compared the results of testing of a group of 53 challenge strains of pneumococci with known resistance properties and a collection of clinical isolates examined in two study phases with a total of 402 and 416 isolates, respectively, with a prototype of the VITEK 2. Testing was conducted in three geographically separate laboratories; the challenge collection was tested by all three laboratories, and the unique clinical isolates were tested separately by the individual laboratories. The VITEK 2 results of tests with 10 antimicrobial agents were compared to the results generated by the National Committee for Clinical Laboratory Standards reference broth microdilution MIC test method. Excellent interlaboratory agreement was observed with the challenge strains. The overall agreement within a single twofold dilution of MICs defined by the VITEK 2 and reference method with the clinical isolates was 96.3%, although there were a number of off-scale MICs that could not be compared. The best agreement with the clinical isolates was achieved with ofloxacin and chloramphenicol (100%), and the lowest level of agreement among those drugs with sufficient on-scale MICs occurred with trimethoprim-sulfamethoxazole (89.7%). Overall there were 1.3% very major, 6.6% minor, and no major interpretive category errors encountered with the clinical isolates, although >80% of the minor interpretive errors involved only a single log(2) dilution difference. The mean time for generation of susceptibility results with the clinical isolates was 8.1 h. The VITEK 2 provided rapid, reliable susceptibility category determinations with both the challenge and clinical isolates examined in this study.