Abstract: Phase-picking models are usually trained on regional waveforms, and their performance on teleseismic records requires further examination using real data. This study takes the DiTing 0.1b phase-picking model of the DiTing seismic large-model series as the study object and carries out a case-based test of teleseismic phase identification within the DTFlowPro framework, using 15 global earthquakes of M5.0～5.9 that occurred from 22 to 31 December 2024. Two detection branches, the native-sampling branch and a 20 Hz resampling branch, were built. The candidates were fused by taking the union of time-clustered picks. The two branches showed clear complementarity. For P phases, 286 stations produced detections in both branches, 179 only in the native-sampling branch, and 40 only in the 20 Hz branch. The corresponding numbers for the S phase were 29, 19, and 25. The fused detections exactly matched the union of the two branches, and the station detection rates increased to 12.03% for the P phase and 1.74% for the S phase, 0.94 and 0.45 percentage points higher than those of the best single branch. The fused P detection rate decreased from 74.19% at 0°～10° to 4.19% at 90°～100°, and the fused S detection rate decreased from 58.06% at 0°～10° to below 8% beyond 10°. Among the combinations of three association methods and three fixed-depth location objective functions, the combination of native sampling, robust residual iteration, and the weighted L2 norm produced the smallest median epicentral error of 23.66 km. The event-building program dtevent of DTFlowPro merged 1102 phase picks and 359 origin solutions into 15 earthquakes, with no missing, incorrectly merged, or incorrectly split catalog events. This case-based evaluation of 15 typical teleseismic earthquakes indicates that, under catalog-guided conditions, the DiTing 0.1b model provides usable identification of teleseismic P phases within 90°, while the identification of S phases and of P phases beyond 90° remains limited. Because these conclusions are based on a limited number of cases, they should not be extrapolated to an assessment of teleseismic monitoring capability.