After the approval of bevacizumab for the first-line treatment of metastatic colorectal cancer (mCRC) in China, published information was still limited. This observational cohort study enrolled 175 mCRC patients who initiated bevacizumab-containing first-line chemotherapies at Sun Yet-sen University Cancer Center. Backbone chemotherapies included FOLFIRI (45.6 %), FOLFOX (34.9 %), and XELOX (19.5 %). Effectiveness data, safety profiles, and treatment patterns were collected and compared between oxaliplatin- and irinotecan-based groups. The median treatment durations of bevacizumab in first-line and total were equivalent between oxaliplatin- and irinotecan-based group (5.0 vs. 4.8 and 6.0 vs. 5.9 months, respectively). Median progression-free survival (PFS) was 10.6 months, and median overall survival (OS) was 24.2 months in entire population. No significant difference was found between irinotecan- and oxaliplatin-based groups in PFS (10.8 vs. 10.1 months, p = 0.21) or in OS (27.5 vs. 23.7 months, p = 0.68). Overall response rate in entire population was 38.3 %, and the disease control rate was 86.3 %. Bevacizumab-associated serious adverse events included hypertension (4.2 %), bleeding (3.6 %), proteinuria (3.0 %), venous thromboembolism (0.6 %), and wound-healing complications (0.6 %). Curative-intent surgery after conversion chemotherapy was carried out in 23 patients (13.7 %). Multivariate analyses showed that maintenance therapy (p = 0.001), resection of metastatic sites (p = 0.002), and disease-free interval (p = 0.003) were independent prognostic factors for OS survival. In spite of small discrepancies in treatment patterns, irinotecan- and oxaliplatin-based chemotherapeutic regimens are equally compatible partners for bevacizumab in first-line Chinese mCRC treatment.