Objective To establish a mouse femoral bone defect model and explore the effect of a GelMA hydrogel on bone regeneration and repair. Methods Forty 8?week?old female C57BL/6N mice were randomly divided into four groups: bone defect group (n= 10), 5% GelMA group (n= 10), 10% GelMA group (n= 10), and 15% GelMA group (n= 10). At 2 weeks after modeling, HE staining was used to observe the femur structure. Masson staining was used to observe the morphology of femoral collagen fibers. OCN immunohistochemical staining was used to analyze bone?specific protein expression levels. Results GelMA was well injectable and injected into bone defects using PCR pipettes. HE staining showed that the 10% GelMA group had a better bone repair effect than bone defect, 5% GelMA, and 15% GelMA groups, and the tissue structure of the defect site was more complete. Masson staining showed more osteogenic collagen fibers in the 10% GelMA group. RT?qPCR analysis showed that OCN expression in the bone defect site in the 10% GelMA group was significantly higher than that in the bone defect group (P< 0?? 001), 5% GelMA group (P< 0?? 01), and 15% GelMA group (P< 0?? 01). Osterix expression in the bone defect site in the 10% GelMA group was significantly higher than that in the bone defect group (P< 0?? 001), 5% GelMA group (P< 0?? 001), and 15% GelMA group (P<0?? 01); OCN immunohistochemical staining showed that bone?specific protein expression in the repair area in the 10% GelMA group was significantly higher than that in the bone defect group (P< 0?? 01), 5% GelMA group (P< 0?? 01), and 15% GelMA group (P< 0?? 05). Conclusions A mouse bone defect model to evaluate the repair effect of injectable regenerative scaffolds was established and applied to the screening and therapeutic effect evaluation of injectable regenerative scaffolds based on GelMA, which provides an experimental basis to carrying out studies in related fields.