Abstract: Generally, porous Ni–Ti shape-memory alloys prepared by the hydrogenation–dehydrogenation process have inferior load-bearing properties and recoverable strains. In this work, high-strength porous Ni–Ti alloys with stabilized cyclic properties were prepared by homogenizing sintering at a high temperature (1250 ℃) under high vacuum conditions (10⁻⁴ Pa) using NaCl as the space holder. High vacuum levels are essential to reduce the risk of sample oxidation during sintering. The sintering process was optimized to ensure the homogenization of the components and densification of the pore wall matrix at 1250 ℃. The alloys with different porosities were studied for their microstructures, phase transformations, mechanical properties, cycle stabilities, and cytotoxicities. Upon increasing the NaCl content from 15% to 40% (volume fraction), the porosities of the samples increased from 14% to 37%, and the average pore size increased from 60 µm to 124 µm, while the oxygen content gradually increased from 0.23% to 0.36% (mass fraction). The porous Ni–Ti alloys predominantly comprised austenite (B2) with a small amount of martensite (B19′) and Ti₂Ni at room temperature (25 ℃). Furthermore, the spontaneous formation of Ni₄Ti₃ nanoprecipitates without heat treatment was observed. The size of the precipitates grew from 20 nm to 145 nm with increasing oxygen content. The martensitic transformation showed multiple peaks in DSC curves attributed to the inhomogeneous distribution of the precipitates. The compressive strengths of the porous Ni–Ti alloys were 1236–1600 MPa. Compared to the porous Ni–Ti alloys prepared by powder metallurgy, the porous Ni–Ti alloys prepared in this study exhibited ultrahigh strength due to matrix strengthening owing to the process optimization. The results of the compression loading–unloading test with 8% strain revealed that the samples exhibited superelasticity as well as shape-memory properties. After heating, the samples’ shape recovery rates exceeded 99%. Under 50 loading–unloading cycles at a constant stress level approaching 8% strain, the irreversible strains of the samples increased with an increasing number of cycles. As the porosity increased, the final residual strains toward the end of the cycle measurements were 1.4%, 1.55%, and 1.66%. These low values of irreversible strains indicated that the porous Ni–Ti samples had excellent cyclic stabilities, which is ascribed to the strengthening effect of Ni₄Ti₃ precipitation in the matrix. To test the cytotoxicity of the porous Ni–Ti alloys, the proliferation of MC3T3E1 cells was tested by the Cell Counting Kit-8 method. The results showed that the cell proliferation rate decreased with increasing porosity, which was due to the release of more Ni ions. Compared to the control group, the proliferation of cells cultured with the Ni–Ti alloys with different porosities in the extracting liquid was optimal. Accordingly, it was shown that the alloys had low cytotoxicity.