Excavated by Seiichi Mizuno in 1965, Lālma is located southwest of the other sites, from which it is separated by a valley. It was chosen for its similarity to the other tepe and because looting threatened to ruin it entirely (Mizuno 1965: 109). Lālma revealed a Stūpas Courtyard and numerous caves. The latter, usually consisting of one room, rarely two, were completely demolished but filled with fragments of modelling and painted plaster.
Built on a shallow terrace accessed by a few steps, like the Great Stūpa M of Tapa-e Shotor, the main stūpa was 8.5m square and had a staircase flanked by two secondary stūpas. At the foot of the platform, on the south-eastern façade to the right of the staircase, were three stūpas. One is attached to it, the others leaving only a very small space between them and the terrace. Similarly, on the northeast side there is a series of eight stūpas. Due to lack of time, the other facades were not excavated. The overflow of the terrace around the Great Stūpa as well as the lack of space between it and the secondary stūpas did not allow for the pradakṣiṇā ritual, which we also observe at Pratès. The decoration of this stūpa was similar to that found at all Haḍḍa sites. Small Buddhas seated in dhyānamudrā were posed against the staircase step on the terrace. The first body of the main stūpa was punctuated by pilasters delineating eight panels. In each of the spaces stood a standing Buddha on a low pedestal flanked by two smaller identical Buddhas. On either side of the staircase, three-lobed niches housed Buddhas in vitarkamudrā. The terrace was similarly decorated with seven Buddhas modellings, either standing with their hands protruding through the sinus of their collars or seated. Numerous Buddhas and Bodhisattvas heads were found in the rubble, all identical to those of the sites already excavated by Barthoux. The excavation was not extensive, so we do not know whether the Stūpas Courtyard was surrounded by an enclosure, whether it was pierced by niches and/or whether it communicated with vihāras or other constructions. The similarities between Lālma and Tapa-e Shotor suggest that they were contemporaneous during a certain period (Kuwayama 1987: 174). Mizuno dates the first floor to the 4th-5th century CE and the second floor to the 5th century.
Plan of Lālma drawn by S. Mizuno (Mizuno, 1968).