Aim: Tuberculosis is a global health problem. To control the disease, timely diagnosis is necessary. Both solid and liquid cultures have been used for diagnosis.  Brochoalveolar lavage is used as a specimen for diagnosis of smear negative tuberculosis and in patients who are   unable to expectorate sputum. The study was done to evaluate the performance for the detection of M. tuberculosis from BAL in terms of detection rates, time to positivity and contamination. Study Design: This was a descriptive cross sectional study. Methodology: BAL samples of suspected patients were collected in the Department of Microbiology, Government medical college, Srinagar. Samples were processed, decontaminated and subjected to smear microscopy by ZN staining, culture on solid (LJ) and liquid media (BacT Alert.). Results: Out of 283 BAL specimens, 23 (8.13%) specimens were AFB smear positive and 260 (91.87%) were AFB smear negative.  The detection rate for Mycobacterium tuberculosis was 8.83% (25/283) on LJ and 12.37% (37/283) on BacT Alert. Overall, detection rate was comparable on smear positive samples but BacT Alert was more sensitive (7.7% detection rate) than LJ (3.46%) in smear-negative specimens. Although detection rates using BacT Alert were greater than LJ, 6 samples were positive exclusively on LJ medium. Time to positivity was significantly lower in liquid media but contamination rates were higher. Conclusion: BacT/Alert affords rapid diagnosis of infection and therefore prompt institution of treatment for M. tuberculosis. Although BacT/Alert has higher sensitivity particularly in smear negative cases, there still remains a proportion of MTB specimens that may go undetected if not complemented with the solid media. Therefore it would be prudent if both the tests are used simultaneously to improve diagnostic yield.