Aims <p>This research examined how the biochar application affects soil aggregates and the associated carbon fractions in tropical agricultural soils through a field experiment.</p> Methods <p>Soil samples were obtained from the 0–20&#xa0;cm soil layer after two years of biochar application to determine the soil aggregate size distribution, stability, soil organic carbon (SOC) and its labile fractions.</p> Results <p>10 t ha⁻¹ biochar application reduced the macroaggregate (&gt; 2&#xa0;mm) content, mean weight diameter (MWD), geometric mean diameter (GMD) and relative stability (R_0.25) but increased the silt + clay aggregate (&lt; 0.053&#xa0;mm) content; ≥ 20 t ha⁻¹ biochar reduced the microaggregate (0.25–0.053&#xa0;mm) or silt + clay aggregate content. Compared with the control treatment, peanut shell biochar (PB) and rice husk biochar (RB) elevated the SOC and total nitrogen (TN) contents of all the aggregates by up to 95% and 62%, respectively, while also enhancing the SOC and TN stocks within the mesoaggregates or microaggregates. Biochar application increased microbial biomass carbon (MBC) and dissolved organic carbon (DOC) contents by 7–259% and 7–36% than the control treatment, respectively. Relative to the control treatment, RB reduced the soil mineralized carbon (SMC) contents of all aggregate sizes. RDA showed that cation exchange capacity (CEC) was the most important factor contributed to soil carbon fractions in aggregates and explained 57% of the variation.</p> Conclusions <p>Biochar application at ≥ 20 t ha⁻¹ is recommended to improve soil aggregate stability, SOC and TN levels, and soil microbial activity in tropical agricultural fields.</p>