Co-pyrolysis can modify biomass product distributions through interactions among feedstock-derived volatiles, minerals, and solid intermediates, yet this behavior remains poorly documented for tropical Indonesian residues. This study investigated ulin wood powder (KU), rubber seed shell (CK), and KU:CK blends of 75:25, 50:50, and 25:75 at 300 °C in a batch fixed-bed reactor. Recovered tar/bio-oil and biochar were measured directly, whereas the apparent gas/unrecovered fraction was calculated by mass balance. Biochar and condensed products were characterized by Fourier-transform infrared spectroscopy (FTIR), headspace gas chromatography-mass spectrometry (GC-MS-HS), and scanning electron microscopy (SEM). KU 100% produced the highest tar/bio-oil yield (207 g; 41.40 wt%), whereas CK 100% produced the highest biochar yield (241 g; 48.20 wt%). The KU:CK 50:50 blend generated the highest apparent gas/unrecovered fraction (205.65 g; 41.13 wt%). Relative to a linear additivity model, the 50:50 blend deviated by -6.93, +0.90, and +6.03 percentage points for tar/bio-oil, biochar, and the apparent gas/unrecovered fraction, respectively; the direction and magnitude of deviations varied with blend ratio. FTIR confirmed the persistence of oxygenated, aliphatic, and aromatic functional-group regions in all biochars, while GC-MS-HS detected volatile oxygenated and aromatic markers, including benzene, toluene, and phenol. These results demonstrate ratio-dependent, non-additive product redistribution at 300 °C without catalyst addition. Because the gas fraction was inferred by difference and replicate-level data were unavailable for statistical reanalysis, the observed non-additivity should be interpreted descriptively rather than as statistically established mechanistic synergy.