Messenger RNA (mRNA)-lipid nanoparticle (mRNA-LNP) platforms enable the in vivo expression of almost any therapeutic protein, offering unprecedented flexibility for clinical translation. However, for these rapidly deployable therapies, predicting the first-in-human (FIH) dose remains a key challenge. We developed an allometric scaling framework for human pharmacokinetic (PK) prediction of antibodies expressed from intravenously administered mRNA-LNPs, leveraging preclinical and clinical data for BNT141 (encoding RiboMab01, a full IgG1) and BNT142 (encoding RiboMab02.1, a bispecific Fab-scFv-based T-cell engager). The dose-normalized Cmax (DCmax) and dose-normalized AUC (DAUC) of translated antibodies across multiple species could be described by the allometric approach, with the estimated exponents ranging from -1.29 to -1.42 for BNT141 and BNT142. We also determined generalized single-species allometric scaling exponents of -1.26 from mice and -0.75 from nonhuman primate (NHP), respectively, that enabled the human predictions of translated antibodies exposure approximately 4-fold (mice) and within 2-fold (NHP). A mechanistic cross-species translational model was further developed that integrated mRNA-specific elimination and translation efficiency parameters to characterize the exposure of the translated antibodies from mRNA-based therapeutics. The translational model could predict the human PK exposure of translated antibodies within 1.33-fold for BNT141 and BNT142 via single-species scaling from NHP parameters as well as capturing the concentration-time profiles with reasonably high fidelity. Validation with publicly available mRNA-1944 (encoding CHKV-24, full IgG1) data confirmed both the allometric scaling and translational modeling methods' robustness when scaling from NHP data. Taken together, these results support a generalized cross-species PK relationship that is independent of the molecular characteristics of the translated antibodies or antibody-like protein products. This integrated scaling and modeling framework offers a generalizable solution for accelerating FIH dose selection of mRNA-encoded therapeutic antibodies and is adaptable to other recombinant proteins of interest.